Organizations face vital challenges in making their recruitment processes extra environment friendly whereas sustaining honest hiring practices. By utilizing AI to remodel their recruitment and expertise acquisition processes, organizations can overcome these challenges. AWS presents a collection of AI providers<\/a> that can be utilized to considerably improve the effectivity, effectiveness, and equity of hiring practices. With AWS AI providers, particularly Amazon Bedrock<\/a>, you’ll be able to construct an environment friendly and scalable recruitment system that streamlines hiring processes, serving to human reviewers give attention to the interview and evaluation of candidates.<\/p>\n On this publish, we present find out how to create an AI-powered recruitment system utilizing Amazon Bedrock, Amazon Bedrock Data Bases<\/a>, AWS Lambda<\/a>, and different AWS providers to boost job description creation, candidate communication, and interview preparation whereas sustaining human oversight.<\/p>\n The recruitment course of presents quite a few alternatives for AI enhancement via specialised brokers<\/a>, every powered by Amazon Bedrock and related to devoted Amazon Bedrock data bases. Let\u2019s discover how these brokers work collectively throughout key levels of the recruitment lifecycle.<\/p>\n Creating inclusive and engaging job descriptions is essential for attracting various expertise swimming pools. The Job Description Creation and Optimization Agent makes use of superior language fashions obtainable in Amazon Bedrock and connects to an Amazon Bedrock data base containing your group\u2019s historic job descriptions and inclusion pointers.<\/p>\n Deploy the Job Description Agent with a safe Amazon Digital Personal Cloud<\/a> (Amazon VPC) configuration and AWS Id and Entry Administration<\/a> (IAM) roles<\/a>. The agent references your data base to optimize job postings whereas sustaining compliance with organizational requirements and inclusive language necessities.<\/p>\n The Candidate Communication Agent manages candidate interactions via the next parts:<\/p>\n Configure the Communication Agent with correct VPC endpoints and encryption for all information in transit and at relaxation. Use Amazon CloudWatch<\/a> monitoring to trace communication effectiveness and response charges.<\/p>\n The Interview Prep Agent helps the interview course of by:<\/p>\n Though the agent gives interview construction and steering, interviewers preserve full management over the dialog and analysis course of.<\/p>\n The structure brings collectively the recruitment brokers and AWS providers right into a complete recruitment system that enhances and streamlines the hiring course of.The next diagram exhibits how three specialised AI brokers work collectively to handle completely different features of the recruitment course of, from job posting creation via summarizing interview suggestions. Every agent makes use of Amazon Bedrock and connects to devoted Amazon Bedrock data bases whereas sustaining safety and compliance necessities.<\/p>\n The answer consists of three predominant parts working collectively to enhance the recruitment course of:<\/p>\n Earlier than implementing this AI-powered recruitment system, be sure to have the next:<\/p>\n Within the following sections, we study the important thing parts that make up our AI-powered recruitment system. Each bit performs a vital position in making a safe, scalable, and efficient resolution. We begin with the infrastructure definition and work our method via the deployment, data base integration, core AI brokers, and testing instruments.<\/p>\n The next AWS CloudFormation<\/a> template defines the whole AWS infrastructure, together with VPC configuration, safety teams, Lambda features, API Gateway, and data bases. It amenities safe, scalable deployment with correct IAM roles and encryption.<\/p>\n The next automation script handles deployment of the recruitment system infrastructure and Lambda features. It manages CloudFormation stack creation and updates and Lambda perform code updates, making system deployment and updates streamlined and constant.<\/p>\n The central data base supervisor interfaces with Amazon Bedrock data base collections to supply greatest practices, templates, and requirements to the recruitment brokers. It permits AI brokers to make knowledgeable choices primarily based on organizational data.<\/p>\n To enhance Retrieval Augmented Technology (RAG) high quality, begin by tuning your Amazon Bedrock data bases. Regulate chunk sizes and overlap in your paperwork, experiment with completely different embedding fashions, and allow reranking to advertise probably the most related passages. For every agent, you can even select completely different basis fashions. For instance, use a quick mannequin reminiscent of Anthropic\u2019s Claude 3 Haiku for high-volume job description and communication duties, and a extra succesful mannequin reminiscent of Anthropic\u2019s Claude 3 Sonnet or one other reasoning-optimized mannequin for the Interview Prep Agent, the place deeper evaluation is required. Seize these experiments as a part of your steady enchancment course of so you’ll be able to standardize on the best-performing configurations.<\/p>\n The mixing between the three brokers is dealt with via API Gateway and Lambda, with every agent uncovered via its personal endpoint. The system makes use of three specialised AI brokers.<\/p>\n This agent is step one within the recruitment pipeline. It makes use of Amazon Bedrock to create inclusive and efficient job descriptions by combining necessities with greatest practices from the data base.<\/p>\n This agent manages candidate communications all through the recruitment course of. It integrates with Amazon SNS for notifications and gives skilled, constant messaging utilizing accepted templates.<\/p>\n This agent prepares tailor-made interview supplies and questions primarily based on the position and candidate background. It helps preserve constant interview requirements whereas adapting to particular positions.<\/p>\n The next take a look at shopper demonstrates interplay with the recruitment system API. It gives instance utilization of main features and helps confirm system performance.<\/p>\n Throughout testing, observe each qualitative and quantitative outcomes. For instance, measure recruiter satisfaction with generated job descriptions, response charges to candidate communications, and interviewers\u2019 suggestions on the usefulness of prep supplies. Use these metrics to refine prompts, data base contents, and mannequin decisions over time.<\/p>\n To keep away from ongoing fees once you\u2019re accomplished testing or if you wish to tear down this resolution, comply with these steps so as:<\/p>\n After cleanup, take these steps to confirm all fees are stopped:<\/p>\n Doc the assets you\u2019ve created and use this record as a guidelines throughout cleanup to be sure to don\u2019t miss any parts that might proceed to generate fees.<\/p>\n To efficiently implement AI in recruitment whereas sustaining moral requirements and human oversight, take into account these important practices.<\/p>\n The safety implementation ought to comply with a complete method to guard all features of the recruitment system. The answer deploys inside a correctly configured VPC with fastidiously outlined safety teams. All information, whether or not at relaxation or in transit, must be protected via AWS KMS encryption, and IAM roles are carried out following strict least privilege rules. The system maintains full visibility via CloudWatch monitoring and audit logging, with safe API Gateway endpoints managing exterior communications. To guard delicate data, implement information tokenization for personally identifiable data (PII) and preserve strict information retention insurance policies. Common privateness impression assessments and documented incident response procedures help ongoing safety compliance.Contemplate the implementation of Amazon Bedrock Guardrails to supply granular management over AI mannequin outputs, serving to you implement constant security and compliance requirements throughout your AI functions. By implementing rule-based filters and bounds, groups can stop inappropriate content material, preserve skilled communication requirements, and ensure responses align with their group\u2019s insurance policies. You possibly can configure guardrails at a number of ranges\u2014from particular person brokers to organization-wide implementations\u2014with customizable controls for content material filtering, subject restrictions, and response parameters. This systematic method helps organizations mitigate dangers whereas utilizing AI capabilities, notably in regulated industries or customer-facing functions the place sustaining applicable, unbiased, and secure interactions is essential.<\/p>\n The data base structure ought to comply with a hub-and-spoke mannequin centered round a core repository of organizational data. This central hub maintains important data together with firm values, insurance policies, and necessities, together with shared reference information used throughout the brokers. Model management and backup procedures preserve information integrity and availability.Surrounding this central hub, specialised data bases serve every agent\u2019s distinctive wants. The Job Description Agent accesses writing pointers and inclusion necessities. The Communication Agent attracts from accepted message templates and workflow definitions, and the Interview Prep Agent makes use of complete query banks and analysis standards.<\/p>\n Profitable system operation depends on sturdy integration practices and clearly outlined workflows. Error dealing with and retry mechanisms facilitate dependable operation, and clear handoff factors between brokers preserve course of integrity. The system ought to preserve detailed documentation of dependencies and information flows, with circuit breakers defending towards cascade failures. Common testing via automated frameworks and end-to-end workflow validation helps constant efficiency and reliability.<\/p>\n The AI-powered recruitment system ought to prioritize human oversight and governance to advertise moral and honest practices. Set up necessary assessment checkpoints all through the method the place human recruiters assess AI suggestions and make closing choices. To deal with distinctive circumstances, create clear escalation paths that enable for human intervention when wanted. Delicate actions, reminiscent of closing candidate picks or provide approvals, must be topic to multi-level human approval workflows.To take care of excessive requirements, repeatedly monitor choice high quality and accuracy, evaluating AI suggestions with human choices to establish areas for enchancment. The staff ought to bear common coaching applications to remain up to date on the system\u2019s capabilities and limitations, ensuring they’ll successfully oversee and complement the AI\u2019s work. Doc clear override procedures, so recruiters can alter or override AI choices when vital. Common compliance coaching for staff members reinforces the dedication to moral AI use in recruitment.<\/p>\n To optimize system effectivity and handle prices successfully, implement a multi-faceted method. Automated scaling for Lambda features makes certain the system can deal with various workloads with out pointless useful resource allocation. For predictable workloads, use AWS Financial savings Plans to cut back prices with out sacrificing efficiency. You possibly can estimate the answer prices utilizing the AWS Pricing Calculator<\/a>, which helps plan for providers like Amazon Bedrock, Lambda, and Amazon Bedrock Data Bases.<\/p>\n Complete CloudWatch dashboards present real-time visibility into system efficiency, facilitating fast identification and addressing of points. Set up efficiency baselines and repeatedly monitor towards these to detect deviations or areas for enchancment. Price allocation tags<\/a> assist observe bills throughout completely different departments or initiatives, enabling extra correct budgeting and useful resource allocation.<\/p>\n To keep away from sudden prices, configure price range alerts that notify the staff when spending approaches predefined thresholds. Common capability planning opinions make certain the infrastructure retains tempo with organizational development and altering recruitment wants.<\/p>\n Dedication to excellence must be mirrored in a steady enchancment framework. Conduct common metric opinions and collect stakeholder suggestions to establish areas for enhancement. A\/B testing of latest options or course of modifications permits for data-driven choices about enhancements. Preserve a complete system of documentation, capturing classes realized from every iteration or problem encountered. This data informs ongoing coaching information updates, ensuring AI fashions stay present and efficient. The development cycle ought to embrace common system optimization, the place algorithms are fine-tuned, data bases up to date, and workflows refined primarily based on efficiency information and person suggestions. Intently analyze efficiency traits over time, permitting proactive addressing of potential points and capitalization on profitable methods. Stakeholder satisfaction must be a key metric within the enchancment framework. Often collect suggestions from recruiters, hiring managers, and candidates to confirm if the AI-powered system meets the wants of all events concerned within the recruitment course of.<\/p>\nThe AI-powered recruitment lifecycle<\/h2>\n
Job description creation and optimization<\/h3>\n
Candidate communication administration<\/h3>\n
\n
Interview preparation and suggestions<\/h3>\n
\n
Answer overview<\/h2>\n
![\"\"](\"https:\/\/d2908q01vomqb2.cloudfront.net\/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59\/2026\/01\/26\/image-1-2.jpg\")
<\/h2>\n\n
Stipulations<\/h2>\n
\n
\n
\n
\n
Infrastructure as code<\/h2>\n
AWSTemplateFormatVersion: '2010-09-09'\nDescription: 'AI-Powered Recruitment System with Safety and Data Bases'\n\nParameters:\n\u00a0\u00a0Atmosphere:\n\u00a0\u00a0 \u00a0Kind: String\n\u00a0\u00a0 \u00a0Default: dev\n\u00a0\u00a0 \u00a0AllowedValues: [dev, prod]\n\nAssets:\n\u00a0\u00a0# KMS Key for encryption\n\u00a0\u00a0RecruitmentKMSKey:\n\u00a0\u00a0 \u00a0Kind: AWS::KMS::Key\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0Description: \"Encryption key for recruitment system\"\n\u00a0\u00a0 \u00a0 \u00a0KeyPolicy:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Assertion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Principal:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0AWS: !Sub 'arn:aws:iam::${AWS::AccountId}:root'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion: 'kms:*'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: '*'\n\n\u00a0\u00a0RecruitmentKMSAlias:\n\u00a0\u00a0 \u00a0Kind: AWS::KMS::Alias\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0AliasName: !Sub 'alias\/recruitment-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0TargetKeyId: !Ref RecruitmentKMSKey\n\n\u00a0\u00a0# VPC Configuration\n\u00a0\u00a0RecruitmentVPC:\n\u00a0\u00a0 \u00a0Kind: AWS::EC2::VPC\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0CidrBlock: 10.0.0.0\/16\n\u00a0\u00a0 \u00a0 \u00a0EnableDnsHostnames: true\n\u00a0\u00a0 \u00a0 \u00a0EnableDnsSupport: true\n\u00a0\u00a0 \u00a0 \u00a0Tags:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- Key: Identify\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Worth: !Sub 'recruitment-vpc-${Atmosphere}'\n\n\u00a0\u00a0PrivateSubnet:\n\u00a0\u00a0 \u00a0Kind: AWS::EC2::Subnet\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0 \u00a0CidrBlock: 10.0.1.0\/24\n\u00a0\u00a0 \u00a0 \u00a0AvailabilityZone: !Choose [0, !GetAZs '']\n\u00a0\n\u00a0PrivateSubnetRouteTable:\n\u00a0\u00a0 \u00a0Kind: AWS::EC2::RouteTable\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0 \u00a0Tags:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- Key: Identify\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Worth: !Sub 'recruitment-private-rt-${Atmosphere}'\n\u00a0\n\u00a0PrivateSubnetRouteTableAssociation:\n\u00a0\u00a0 \u00a0Kind: AWS::EC2::SubnetRouteTableAssociation\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0SubnetId: !Ref PrivateSubnet\n\u00a0\u00a0 \u00a0 \u00a0RouteTableId: !Ref PrivateSubnetRouteTable\n\u00a0\n# Instance Interface Endpoints\nVPCEBedrockRuntime:\n\u00a0\u00a0Kind: AWS::EC2::VPCEndpoint\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0ServiceName: !Sub 'com.amazonaws.${AWS::Area}.bedrock-runtime'\n\u00a0\u00a0 \u00a0VpcEndpointType: Interface\n\u00a0\u00a0 \u00a0SubnetIds: [ !Ref PrivateSubnet ]\n\u00a0\u00a0 \u00a0SecurityGroupIds: [ !Ref LambdaSecurityGroup ]\n\nVPCEBedrockAgent:\n\u00a0\u00a0Kind: AWS::EC2::VPCEndpoint\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0ServiceName: !Sub 'com.amazonaws.${AWS::Area}.bedrock-agent'\n\u00a0\u00a0 \u00a0VpcEndpointType: Interface\n\u00a0\u00a0 \u00a0SubnetIds: [ !Ref PrivateSubnet ]\n\u00a0\u00a0 \u00a0SecurityGroupIds: [ !Ref LambdaSecurityGroup ]\n\nVPCESNS:\n\u00a0\u00a0Kind: AWS::EC2::VPCEndpoint\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0ServiceName: !Sub 'com.amazonaws.${AWS::Area}.sns'\n\u00a0\u00a0 \u00a0VpcEndpointType: Interface\n\u00a0\u00a0 \u00a0SubnetIds: [ !Ref PrivateSubnet ]\n\u00a0\u00a0 \u00a0SecurityGroupIds: [ !Ref LambdaSecurityGroup ]\n\n# Gateway endpoints for S3 (and DynamoDB in case you add it later)\nVPCES3:\n\u00a0\u00a0Kind: AWS::EC2::VPCEndpoint\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0ServiceName: !Sub 'com.amazonaws.${AWS::Area}.s3'\n\u00a0\u00a0 \u00a0VpcEndpointType: Gateway\n\u00a0\u00a0 \u00a0RouteTableIds:\n\u00a0\u00a0 \u00a0 \u00a0- !Ref PrivateSubnetRouteTable \u00a0 # create if not current\n\u00a0\u00a0# Safety Group\n\u00a0\u00a0LambdaSecurityGroup:\n\u00a0\u00a0 \u00a0Kind: AWS::EC2::SecurityGroup\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0GroupDescription: Safety group for recruitment AWS\u00a0Lambda features\n\u00a0\u00a0 \u00a0 \u00a0VpcId: !Ref RecruitmentVPC\n\u00a0\u00a0 \u00a0 \u00a0SecurityGroupEgress:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- IpProtocol: tcp\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0FromPort: 443\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0ToPort: 443\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0CidrIp: 0.0.0.0\/0\n\n\u00a0 #\u00a0KnowledgeBase IAM position\n\u00a0\u00a0KnowledgeBaseRole:\n\u00a0\u00a0Kind: AWS::IAM::Function\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0AssumeRolePolicyDocument:\n\u00a0\u00a0 \u00a0 \u00a0Model: '2012-10-17'\n\u00a0\u00a0 \u00a0 \u00a0Assertion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Principal: { Service: bedrock.amazonaws.com }\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion: sts:AssumeRole\n\u00a0\u00a0 \u00a0Insurance policies:\n\u00a0\u00a0 \u00a0 \u00a0- PolicyName: BedrockKBAccess\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0PolicyDocument:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Model: '2012-10-17'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0Assertion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- bedrock:Retrieve\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- bedrock:RetrieveAndGenerate\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: \"*\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- s3:GetObject\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- s3:ListBucket\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: \"*\" \u00a0 # scope to your KB bucket(s) in actual deployments\n\n\u00a0 \u00a0 JobDescriptionKnowledgeBase:\n\u00a0 \u00a0 \u00a0 \u00a0\u00a0Kind: AWS::Bedrock::KnowledgeBase\n\u00a0 \u00a0 \u00a0 \u00a0\u00a0Properties:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0Identify: !Sub 'job-descriptions-${Atmosphere}'\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0RoleArn: !GetAtt KnowledgeBaseRole.Arn\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0KnowledgeBaseConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0Kind: VECTOR\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0VectorKnowledgeBaseConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0EmbeddingModelArn: !Sub 'arn:aws:bedrock:${AWS::Area}::foundation-model\/amazon.titan-embed-text-v1'\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0StorageConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0Kind: S3\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0S3Configuration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0BucketArn: !Sub 'arn:aws:s3:::your-kb-bucket-${Atmosphere}-${AWS::AccountId}-${AWS::Area}'\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0BucketOwnerAccountId: !Ref AWS::AccountId\n\n\u00a0 \u00a0 InterviewKnowledgeBase:\n\u00a0 \u00a0 \u00a0 \u00a0\u00a0Kind: AWS::Bedrock::KnowledgeBase\n\u00a0 \u00a0 \u00a0 \u00a0\u00a0Properties:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0Identify: !Sub 'interview-standards-${Atmosphere}'\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0RoleArn: !GetAtt KnowledgeBaseRole.Arn\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0KnowledgeBaseConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0Kind: VECTOR\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0VectorKnowledgeBaseConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 EmbeddingModelArn: arn:aws:bedrock:${AWS::Area}::foundation-model\/amazon.titan-embed-text-v2:0\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0StorageConfiguration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0Kind: S3\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0S3Configuration:\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0BucketArn: !Sub 'arn:aws:s3:::your-kb-bucket-${Atmosphere}-${AWS::AccountId}-${AWS::Area}'\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0BucketOwnerAccountId: !Ref AWS::AccountId\n\n\u00a0\u00a0# CloudTrail for audit logging\n\u00a0\u00a0RecruitmentCloudTrail:\n\u00a0\u00a0 \u00a0Kind: AWS::CloudTrail::Path\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0TrailName: !Sub 'recruitment-audit-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0S3BucketName: !Ref AuditLogsBucket\n\u00a0\u00a0 \u00a0 \u00a0IncludeGlobalServiceEvents: true\n\u00a0\u00a0 \u00a0 \u00a0IsMultiRegionTrail: true\n\u00a0\u00a0 \u00a0 \u00a0EnableLogFileValidation: true\n\u00a0\u00a0 \u00a0 \u00a0KMSKeyId: !Ref RecruitmentKMSKey\n\n\u00a0\u00a0AuditLogsBucket:\n\u00a0\u00a0 \u00a0Kind: AWS::S3::Bucket\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0BucketName: !Sub 'recruitment-audit-logs-${Atmosphere}-${AWS::AccountId}-${AWS::Area}'\n\u00a0\u00a0 \u00a0 \u00a0BucketEncryption:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0ServerSideEncryptionConfiguration:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0- ServerSideEncryptionByDefault:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0SSEAlgorithm: aws:kms\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0KMSMasterKeyID: !Ref RecruitmentKMSKey\n\u00a0\u00a0# IAM Function for AWS Lambda features\n\u00a0\u00a0LambdaExecutionRole:\n\u00a0\u00a0 \u00a0Kind: AWS::IAM::Function\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0AssumeRolePolicyDocument:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Model: '2012-10-17'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Assertion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Principal:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Service: lambda.amazonaws.com\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion: sts:AssumeRole\n\u00a0\u00a0 \u00a0 \u00a0ManagedPolicyArns:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- arn:aws:iam::aws:coverage\/service-role\/AWSLambdaBasicExecutionRole\n\u00a0\u00a0 \u00a0 \u00a0Insurance policies:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0- PolicyName: BedrockAccess\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0PolicyDocument:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Model: '2012-10-17'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Assertion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- bedrock:InvokeModel\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- bedrock:Retrieve\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: '*'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- sns:Publish\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: !Ref CommunicationTopic\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- kms:Decrypt\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- kms:GenerateDataKey\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: !GetAtt RecruitmentKMSKey.Arn\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- Impact: Permit\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Motion:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0- aoss:APIAccessAll\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Useful resource: '*'\n\n\u00a0\u00a0# SNS Matter for notifications\n\u00a0\u00a0CommunicationTopic:\n\u00a0\u00a0 \u00a0Kind: AWS::SNS::Matter\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0TopicName: !Sub 'recruitment-notifications-${Atmosphere}'\n\n\u00a0\u00a0# AWS Lambda Capabilities\n\u00a0\u00a0JobDescriptionFunction:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Perform\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Sub 'recruitment-job-description-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0Runtime: python3.11\n\u00a0\u00a0 \u00a0 \u00a0Handler: job_description_agent.lambda_handler\n\u00a0\u00a0 \u00a0 \u00a0Function: !GetAtt LambdaExecutionRole.Arn\n\u00a0\u00a0 \u00a0 \u00a0Code:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0ZipFile: |\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0# Code will probably be deployed individually\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0def lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return {'statusCode': 200, 'physique': 'Placeholder'}\n\u00a0\u00a0 \u00a0 \u00a0Timeout: 60\n\n\u00a0\u00a0CommunicationFunction:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Perform\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Sub 'recruitment-communication-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0Runtime: python3.11\n\u00a0\u00a0 \u00a0 \u00a0Handler: communication_agent.lambda_handler\n\u00a0\u00a0 \u00a0 \u00a0Function: !GetAtt LambdaExecutionRole.Arn\n\u00a0\u00a0 \u00a0 \u00a0Code:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0ZipFile: |\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0def lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return {'statusCode': 200, 'physique': 'Placeholder'}\n\u00a0\u00a0 \u00a0 \u00a0Timeout: 60\n\u00a0\u00a0 \u00a0 \u00a0Atmosphere:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Variables:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0SNS_TOPIC_ARN: !Ref CommunicationTopic\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0KMS_KEY_ID: !Ref RecruitmentKMSKey\n\u00a0\u00a0 \u00a0 \u00a0VpcConfig:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0SecurityGroupIds:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0- !Ref LambdaSecurityGroup\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0SubnetIds:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0- !Ref PrivateSubnet\n\n\u00a0\u00a0InterviewFunction:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Perform\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Sub 'recruitment-interview-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0Runtime: python3.11\n\u00a0\u00a0 \u00a0 \u00a0Handler: interview_agent.lambda_handler\n\u00a0\u00a0 \u00a0 \u00a0Function: !GetAtt LambdaExecutionRole.Arn\n\u00a0\u00a0 \u00a0 \u00a0Code:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0ZipFile: |\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0def lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return {'statusCode': 200, 'physique': 'Placeholder'}\n\u00a0\u00a0 \u00a0 \u00a0Timeout: 60\n\n\u00a0\u00a0# API Gateway\n\u00a0\u00a0RecruitmentAPI:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::RestApi\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0Identify: !Sub 'recruitment-api-${Atmosphere}'\n\u00a0\u00a0 \u00a0 \u00a0Description: 'API for AI-Powered Recruitment System'\n\n\u00a0\u00a0# API Gateway Assets and Strategies\n\u00a0\u00a0JobDescriptionResource:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Useful resource\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ParentId: !GetAtt RecruitmentAPI.RootResourceId\n\u00a0\u00a0 \u00a0 \u00a0PathPart: job-description\n\n\u00a0\u00a0JobDescriptionMethod:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Methodology\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ResourceId: !Ref JobDescriptionResource\n\u00a0\u00a0 \u00a0 \u00a0HttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0AuthorizationType: NONE\n\u00a0\u00a0 \u00a0 \u00a0Integration:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Kind: AWS_PROXY\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0IntegrationHttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Uri: !Sub 'arn:aws:apigateway:${AWS::Area}:lambda:path\/2015-03-31\/features\/${JobDescriptionFunction.Arn}\/invocations'\n\n\u00a0 CommunicationResource:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Useful resource\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ParentId: !GetAtt RecruitmentAPI.RootResourceId\n\u00a0\u00a0 \u00a0 \u00a0PathPart: communication\n\n\u00a0 CommunicationMethod:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Methodology\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ResourceId: !Ref CommunicationResource\n\u00a0\u00a0 \u00a0 \u00a0HttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0AuthorizationType: NONE\n\u00a0\u00a0 \u00a0 \u00a0Integration:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Kind: AWS_PROXY\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0IntegrationHttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Uri: !Sub 'arn:aws:apigateway:${AWS::Area}:lambda:path\/2015-03-31\/features\/${CommunicationFunction.Arn}\/invocations'\n\n\u00a0 InterviewResource:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Useful resource\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ParentId: !GetAtt RecruitmentAPI.RootResourceId\n\u00a0\u00a0 \u00a0 \u00a0PathPart: interview\n\n\u00a0 InterviewMethod:\n\u00a0\u00a0 \u00a0Kind: AWS::ApiGateway::Methodology\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0 \u00a0ResourceId: !Ref InterviewResource\n\u00a0\u00a0 \u00a0 \u00a0HttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0AuthorizationType: NONE\n\u00a0\u00a0 \u00a0 \u00a0Integration:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Kind: AWS_PROXY\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0IntegrationHttpMethod: POST\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0Uri: !Sub 'arn:aws:apigateway:${AWS::Area}:lambda:path\/2015-03-31\/features\/${InterviewFunction.Arn}\/invocations'\n\n\u00a0\u00a0# Lambda Permissions\n\u00a0\u00a0JobDescriptionPermission:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Permission\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Ref JobDescriptionFunction\n\u00a0\u00a0 \u00a0 \u00a0Motion: lambda:InvokeFunction\n\u00a0\u00a0 \u00a0 \u00a0Principal: apigateway.amazonaws.com\n\u00a0\u00a0 \u00a0 \u00a0SourceArn: !Sub '${RecruitmentAPI}\/*\/POST\/job-description'\n\n\u00a0 CommunicationPermission:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Permission\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Ref CommunicationFunction\n\u00a0\u00a0 \u00a0 \u00a0Motion: lambda:InvokeFunction\n\u00a0\u00a0 \u00a0 \u00a0Principal: apigateway.amazonaws.com\n\u00a0\u00a0 \u00a0 \u00a0SourceArn: !Sub '${RecruitmentAPI}\/*\/POST\/communication'\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\n\u00a0 InterviewPermission:\n\u00a0\u00a0 \u00a0Kind: AWS::Lambda::Permission\n\u00a0\u00a0 \u00a0Properties:\n\u00a0\u00a0 \u00a0 \u00a0FunctionName: !Ref InterviewFunction\n\u00a0\u00a0 \u00a0 \u00a0Motion: lambda:InvokeFunction\n\u00a0\u00a0 \u00a0 \u00a0Principal: apigateway.amazonaws.com\n\u00a0\u00a0 \u00a0 \u00a0SourceArn: !Sub '${RecruitmentAPI}\/*\/POST\/interview'\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\n\u00a0\u00a0# API Deployment\n\u00a0\u00a0APIDeployment:\n\u00a0\u00a0Kind: AWS::ApiGateway::Deployment\n\u00a0\u00a0DependsOn:\n\u00a0\u00a0 \u00a0- JobDescriptionMethod\n\u00a0\u00a0 \u00a0- CommunicationMethod\n\u00a0\u00a0 \u00a0- InterviewMethod\n\u00a0\u00a0 \u00a0- JobDescriptionPermission\n\u00a0\u00a0 \u00a0- CommunicationPermission\n\u00a0\u00a0 \u00a0- InterviewPermission\n\u00a0\u00a0Properties:\n\u00a0\u00a0 \u00a0RestApiId: !Ref RecruitmentAPI\n\u00a0\u00a0 \u00a0StageName: !Ref Atmosphere\n\u00a0\nOutputs:\n\u00a0\u00a0APIEndpoint:\n\u00a0\u00a0 \u00a0Description: 'API Gateway endpoint URL'\n\u00a0\u00a0 \u00a0Worth: !Sub 'https:\/\/${RecruitmentAPI}.execute-api.${AWS::Area}.amazonaws.com\/${Atmosphere}'\n\u00a0\u00a0\n\u00a0\u00a0SNSTopicArn:\n\u00a0\u00a0 \u00a0Description: 'SNS Matter ARN for notifications'\n\u00a0\u00a0 \u00a0Worth: !Ref CommunicationTopic<\/code><\/pre>\n<\/p><\/div>\nDeployment automation<\/h2>\n
#!\/usr\/bin\/env python3\n\"\"\"\nDeployment script for Primary Recruitment System\n\"\"\"\n\nimport boto3\nimport zipfile\nimport os\nimport json\nfrom pathlib import Path\n\nclass BasicRecruitmentDeployment:\n\u00a0\u00a0 \u00a0def __init__(self, area='us-east-1'):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.area = area\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.lambda_client = boto3.shopper('lambda', region_name=area)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.cf_client = boto3.shopper('cloudformation', region_name=area)\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0def create_lambda_zip(self, function_name):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\"\"\"Create deployment zip for Lambda perform\"\"\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0zip_path = f\"\/tmp\/{function_name}.zip\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0with zipfile.ZipFile(zip_path, 'w') as zip_file:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0zip_file.write(f\"lambda_functions\/{function_name}.py\", f\"{function_name}.py\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return zip_path\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0def update_lambda_function(self, function_name, surroundings=\"dev\"):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\"\"\"Replace Lambda perform code\"\"\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0zip_path = self.create_lambda_zip(function_name)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0with open(zip_path, 'rb') as zip_file:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0response = self.lambda_client.update_function_code(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0FunctionName=f'recruitment-{function_name.change(\"_agent\", \"\")}-{surroundings}',\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0ZipFile=zip_file.learn()\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0print(f\"Up to date {function_name}: {response['LastModified']}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return response\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0print(f\"Error updating {function_name}: {e}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return None\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0lastly:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0os.take away(zip_path)\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0def deploy_infrastructure(self, surroundings=\"dev\"):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\"\"\"Deploy CloudFormation stack\"\"\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0stack_name = f'recruitment-system-{surroundings}'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0with open('infrastructure\/cloudformation.yaml', 'r') as template_file:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0template_body = template_file.learn()\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0response = self.cf_client.create_stack(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0StackName=stack_name,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0TemplateBody=template_body,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Parameters=[\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0{'ParameterKey': 'Environment', 'ParameterValue': environment}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Capabilities=['CAPABILITY_IAM']\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0print(f\"Created stack: {stack_name}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return response\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0besides self.cf_client.exceptions.AlreadyExistsException:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0response = self.cf_client.update_stack(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0StackName=stack_name,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0TemplateBody=template_body,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Parameters=[\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0{'ParameterKey': 'Environment', 'ParameterValue': environment}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Capabilities=['CAPABILITY_IAM']\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0print(f\"Up to date stack: {stack_name}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return response\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0print(f\"Error with stack: {e}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return None\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0def deploy_all(self, surroundings=\"dev\"):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\"\"\"Deploy full system\"\"\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0print(f\"Deploying recruitment system to {surroundings}\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0# Deploy infrastructure\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.deploy_infrastructure(surroundings)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0# Anticipate stack to be prepared (simplified)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0print(\"Ready for infrastructure...\")\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0# Replace AWS Lambda features\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0features = [\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'job_description_agent',\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'communication_agent',\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'interview_agent'\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0for func in features:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0self.update_lambda_function(func, surroundings)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0print(\"Deployment full!\")\n\ndef predominant():\n\u00a0\u00a0 \u00a0deployment = BasicRecruitmentDeployment()\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0print(\"Primary Recruitment System Deployment\")\n\u00a0\u00a0 \u00a0print(\"1. Deploys CloudFormation stack with AWS Lambda features and API Gateway\")\n\u00a0\u00a0 \u00a0print(\"2. Updates Lambda perform code\")\n\u00a0\u00a0 \u00a0print(\"3. Units up SNS for notifications\")\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0# Instance deployment\n\u00a0\u00a0 \u00a0# deployment.deploy_all('dev')\n\nif __name__ == \"__main__\":\n\u00a0\u00a0 \u00a0predominant()<\/code><\/pre>\n<\/p><\/div>\nData base integration<\/h2>\n
import boto3\nimport json\n\nclass KnowledgeBaseManager:\n\u00a0\u00a0 \u00a0def __init__(self):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.bedrock_runtime = boto3.shopper('bedrock-runtime')\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.bedrock_agent_runtime = boto3.shopper('bedrock-agent-runtime')\n\n\u00a0\u00a0 \u00a0def query_knowledge_base(self, kb_id: str, question: str):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0response = self.bedrock_agent_runtime.retrieve(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0knowledgeBaseId=kb_id,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0retrievalQuery={'textual content': question}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0# optionally add retrievalConfiguration={...}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return [r['content']['text'] for r in response.get('retrievalResults', [])]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0return [f\"Knowledge Base query failed: {str(e)}\"]\n\n# Data base IDs (to be created through CloudFormation)\nKNOWLEDGE_BASES = {\n\u00a0\u00a0 \u00a0'job_descriptions': 'JOB_DESC_KB_ID',\u00a0\n\u00a0\u00a0 \u00a0'interview_standards': 'INTERVIEW_KB_ID',\n\u00a0\u00a0 \u00a0'communication_templates': 'COMM_KB_ID'\n}<\/code><\/pre>\n<\/p><\/div>\nThe core AI brokers<\/h2>\n
Job Description Agent<\/h3>\n
import json\nimport boto3\nfrom datetime import datetime\nimport sys\nimport os\nsys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))\nfrom knowledge_bases import KnowledgeBaseManager, KNOWLEDGE_BASES\n\nbedrock = boto3.shopper('bedrock-runtime')\nkb_manager = KnowledgeBaseManager()\n\ndef lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0\"\"\"Job Description Agent Lambda perform\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0physique = json.masses(occasion.get('physique', '{}'))\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0role_title = physique.get('role_title', '')\n\u00a0\u00a0 \u00a0necessities = physique.get('necessities', [])\n\u00a0\u00a0 \u00a0company_info = physique.get('company_info', {})\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0# Question data base for greatest practices\n\u00a0\u00a0 \u00a0kb_context = kb_manager.query_knowledge_base(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0KNOWLEDGE_BASES['job_descriptions'],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0f\"inclusive job description examples for {role_title}\"\n\u00a0\u00a0 \u00a0)\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0immediate = f\"\"\"Create an inclusive job description for: {role_title}\n\u00a0\u00a0 \u00a0\nNecessities: {', '.be a part of(necessities)}\nFirm: {company_info.get('identify', 'Our Firm')}\nTradition: {company_info.get('tradition', 'collaborative')}\nDistant: {company_info.get('distant', False)}\n\nGreatest practices from data base:\n{' '.be a part of(kb_context[:2])}\n\nEmbrace: position abstract, key duties, {qualifications}, advantages.\nGuarantee inclusive language and keep away from pointless boundaries.\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0response = bedrock.invoke_model(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0modelId=\"anthropic.claude-3-haiku-20240307-v1:0\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0physique=json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"anthropic_version\": \"bedrock-2023-05-31\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"max_tokens\": 2000,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"messages\": [{\"role\": \"user\", \"content\": prompt}]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0outcome = json.masses(response['body'].learn())\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 200,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'headers': {'Content material-Kind': 'utility\/json'},\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'job_description': outcome['content'][0]['text'],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'role_title': role_title,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'timestamp': datetime.utcnow().isoformat()\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 500,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({'error': str(e)})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}<\/code><\/pre>\n<\/p><\/div>\nCommunication Agent<\/h3>\n
import json\nimport boto3\nfrom datetime import datetime\n\nbedrock = boto3.shopper('bedrock-runtime')\nsns = boto3.shopper('sns')\n\ndef lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0\"\"\"Communication Agent Lambda perform\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0physique = json.masses(occasion.get('physique', '{}'))\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0message_type = physique.get('message_type', '')\n\u00a0\u00a0 \u00a0candidate_info = physique.get('candidate_info', {})\n\u00a0\u00a0 \u00a0stage = physique.get('stage', '')\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0immediate = f\"\"\"Generate {message_type} for candidate {candidate_info.get('identify', 'Candidate')} \nat {stage} stage.\n\nMessage must be:\n- Skilled and empathetic\n- Clear about subsequent steps\n- Acceptable for the stage\n- Embrace timeline if related\n\nVarieties: application_received, interview_invitation, rejection, provide\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0response = bedrock.invoke_model(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0modelId=\"anthropic.claude-3-haiku-20240307-v1:0\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0physique=json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"anthropic_version\": \"bedrock-2023-05-31\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"max_tokens\": 1000,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"messages\": [{\"role\": \"user\", \"content\": prompt}]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0outcome = json.masses(response['body'].learn())\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0communication = outcome['content'][0]['text']\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0# Ship notification through SNS if subject ARN supplied\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0topic_arn = physique.get('sns_topic_arn')\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0if topic_arn:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0sns.publish(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0TopicArn=topic_arn,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Message=communication,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Topic=f\"Recruitment Replace - {message_type}\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 200,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'headers': {'Content material-Kind': 'utility\/json'},\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'communication': communication,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'kind': message_type,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'stage': stage,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'timestamp': datetime.utcnow().isoformat()\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 500,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({'error': str(e)})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}<\/code><\/pre>\n<\/p><\/div>\nInterview Prep Agent<\/h3>\n
import json\nimport boto3\nfrom datetime import datetime\n\nbedrock = boto3.shopper('bedrock-runtime')\n\ndef lambda_handler(occasion, context):\n\u00a0\u00a0 \u00a0\"\"\"Interview Prep Agent Lambda perform\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0physique = json.masses(occasion.get('physique', '{}'))\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0role_info = physique.get('role_info', {})\n\u00a0\u00a0 \u00a0candidate_background = physique.get('candidate_background', {})\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0immediate = f\"\"\"Put together interview for:\nFunction: {role_info.get('title', 'Place')}\nStage: {role_info.get('degree', 'Mid-level')}\nKey Expertise: {role_info.get('key_skills', [])}\n\nCandidate Background:\nExpertise: {candidate_background.get('expertise', 'Not specified')}\nExpertise: {candidate_background.get('expertise', [])}\n\nGenerate:\n1. 5-7 technical questions\n2. 3-4 behavioral questions \u00a0\n3. Analysis standards\n4. Crimson flags to look at for\"\"\"\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0attempt:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0response = bedrock.invoke_model(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0modelId=\"anthropic.claude-3-haiku-20240307-v1:0\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0physique=json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"anthropic_version\": \"bedrock-2023-05-31\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"max_tokens\": 2000,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"messages\": [{\"role\": \"user\", \"content\": prompt}]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0outcome = json.masses(response['body'].learn())\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 200,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'headers': {'Content material-Kind': 'utility\/json'},\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'interview_prep': outcome['content'][0]['text'],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'position': role_info.get('title'),\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'timestamp': datetime.utcnow().isoformat()\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0besides Exception as e:\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'statusCode': 500,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0'physique': json.dumps({'error': str(e)})\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}<\/code><\/pre>\n<\/p><\/div>\nTesting and verification<\/h2>\n
#!\/usr\/bin\/env python3\n\"\"\"\nTake a look at shopper for Primary Recruitment System API\n\"\"\"\n\nimport requests\nimport json\n\nclass RecruitmentClient:\n\u00a0\u00a0 \u00a0def __init__(self, api_endpoint):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0self.api_endpoint = api_endpoint.rstrip('\/')\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0def create_job_description(self, role_title, necessities, company_info):\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\"\"\"Take a look at job description creation\"\"\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0url = f\"{self.api_endpoint}\/job-description\"\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0payload = {\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"role_title\": role_title,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"necessities\": necessities,\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"company_info\": company_info\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0response = requests.publish(url, json=payload)\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0return response.json()\n\u00a0 \u00a0\n\u00a0 \u00a0\u00a0def send_communication(self, message_type, candidate_info, stage):\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0\"\"\"Take a look at communication sending\"\"\"\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0url = f\"{self.api_endpoint}\/communication\"\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0payload = {\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0\"message_type\": message_type,\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0\"candidate_info\": candidate_info,\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0\"stage\": stage\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0}\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0response = requests.publish(url, json=payload)\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0return response.json()\n\n\u00a0\u00a0 \u00a0def prepare_interview(self, role_info, candidate_background):\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0\"\"\"Take a look at interview preparation\"\"\"\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0url = f\"{self.api_endpoint}\/interview\"\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0payload = {\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0\"role_info\": role_info,\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0 \u00a0 \u00a0\"candidate_background\": candidate_background\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0}\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0response = requests.publish(url, json=payload)\n\u00a0 \u00a0 \u00a0\u00a0 \u00a0return response.json()\n\ndef predominant():\n\u00a0\u00a0 \u00a0# Change together with your precise API endpoint\n\u00a0\u00a0 \u00a0api_endpoint = \"https:\/\/your-api-id.execute-api.us-east-1.amazonaws.com\/dev\"\n\u00a0\u00a0 \u00a0shopper = RecruitmentClient(api_endpoint)\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0print(\"Testing Primary Recruitment System\")\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0# Take a look at job description\n\u00a0\u00a0 \u00a0print(\"n1. Testing Job Description Creation:\")\n\u00a0\u00a0 \u00a0job_result = shopper.create_job_description(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0role_title=\"Senior Software program Engineer\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0necessities=[\"5+ years Python\", \"AWS experience\", \"Team leadership\"],\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0company_info={\"identify\": \"TechCorp\", \"tradition\": \"collaborative\", \"distant\": True}\n\u00a0\u00a0 \u00a0)\n\u00a0\u00a0 \u00a0print(json.dumps(job_result, indent=2))\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0# Take a look at communication\n\u00a0\u00a0 \u00a0print(\"n2. Testing Communication:\")\n\u00a0\u00a0 \u00a0comm_result = shopper.send_communication(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0message_type=\"interview_invitation\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0candidate_info={\"identify\": \"Jane Smith\", \"electronic mail\": \"jane@instance.com\"},\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0stage=\"initial_interview\"\n\u00a0\u00a0 \u00a0)\n\u00a0\u00a0 \u00a0print(json.dumps(comm_result, indent=2))\n\u00a0\u00a0 \u00a0\n\u00a0\u00a0 \u00a0# Take a look at interview prep\n\u00a0\u00a0 \u00a0print(\"n3. Testing Interview Preparation:\")\n\u00a0\u00a0 \u00a0interview_result = shopper.prepare_interview(\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0role_info={\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"title\": \"Senior Software program Engineer\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"degree\": \"Senior\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"key_skills\": [\"Python\", \"AWS\", \"Leadership\"]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0},\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0candidate_background={\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"expertise\": \"8 years software program improvement\",\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0\"expertise\": [\"Python\", \"AWS\", \"Team Lead\"]\n\u00a0\u00a0 \u00a0 \u00a0 \u00a0}\n\u00a0\u00a0 \u00a0)\n\u00a0\u00a0 \u00a0print(json.dumps(interview_result, indent=2))\n\nif __name__ == \"__main__\":\n\u00a0\u00a0 \u00a0predominant()<\/code><\/pre>\n<\/p><\/div>\nClear up<\/h2>\n
\n
\n
\n
\n
\n
\n
\n
\n
\n
\n
Implementing AI in recruitment: Greatest practices<\/h2>\n
Safety, compliance, and infrastructure<\/h3>\n
Data base structure and administration<\/h3>\n
System integration and workflows<\/h3>\n
Human oversight and governance<\/h3>\n
Efficiency and price administration<\/h3>\n
Steady enchancment framework<\/h3>\n
Answer evolution and agent orchestration<\/h3>\n
![\"\"](\"https:\/\/d2908q01vomqb2.cloudfront.net\/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59\/2026\/02\/04\/Adesanya.jpeg\")
Dola Adesanya<\/strong> is a Buyer Options Supervisor at Amazon Net Providers (AWS), the place she leads high-impact applications throughout buyer success, cloud transformation, and AI-driven system supply. With a singular mix of enterprise technique and organizational psychology experience, she focuses on turning complicated challenges into actionable options. Dola brings in depth expertise in scaling applications and delivering measurable enterprise outcomes.<\/p>\n![\"\"](\"https:\/\/d2908q01vomqb2.cloudfront.net\/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59\/2026\/02\/04\/ronhayman.png\")
Ron Hayman<\/strong> leads Buyer Options for US Enterprise and Software program Web & Basis Fashions at Amazon Net Providers (AWS). His group helps clients migrate infrastructure, modernize functions, and implement generative AI options. Over his 20-year profession as a world know-how government, Ron has constructed and scaled cloud, safety, and buyer success groups. He combines deep technical experience with a confirmed observe document of growing leaders, organizing groups, and delivering buyer outcomes.<\/p>\n![\"\"](\"https:\/\/d2908q01vomqb2.cloudfront.net\/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59\/2026\/02\/04\/achf.jpeg\")
Achilles Figueiredo<\/strong> is a Senior Options Architect at Amazon Net Providers (AWS), the place he designs and implements enterprise-scale cloud architectures. As a trusted technical advisor, he helps organizations navigate complicated digital transformations whereas implementing revolutionary cloud options. He actively contributes to AWS\u2019s technical development via AI, Safety, and Resilience initiatives and serves as a key useful resource for each strategic planning and hands-on implementation steering.<\/p>\n![\"\"](\"https:\/\/d2908q01vomqb2.cloudfront.net\/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59\/2026\/02\/04\/sai.png\")
Sai Jeedigunta<\/strong> is a Sr. Buyer Options Supervisor at AWS. He’s enthusiastic about partnering with executives and cross-functional groups in driving cloud transformation initiatives and serving to them understand the advantages of cloud. He has over 20 years of expertise in main IT infrastructure engagements for fortune enterprises.<\/p>\n