AWS AI Engineer / Developer with AWS Fargate, Bedrock, OpenSearch & Connect Exp. (G.C / U.S.C)
6+Months
Lansing, MI (Onsite Preferred / Remote)
-It can be remote, preference to onsite/Local to MI
W2 Only (No C2C)
Summary:
We are seeking an AWS-focused developer to build and operate agentic AI solutions at scale. You will design LLM-driven agents, secure APIs, and retrieval pipelines on AWS services including Bedrock, Lambda, Fargate, OpenSearch, and Amazon Connect. The ideal candidate blends advanced prompt engineering and Python-based LLM orchestration with solid AWS serverless and container experience.
Key Responsibilities:
- Design, implement, and optimize agentic AI workflows and multi-agent patterns for Q&A, task automation, and contact-center use cases
- Build Python services and REST APIs (FastAPI) that orchestrate LLMs, tools, and data sources
- Implement RAG pipelines using OpenSearch (indexing, embeddings, relevance tuning) and evaluate with objective metrics
- Develop Bedrock-based chatbots and Q&A assistants with guardrails, prompt templates, and structured outputs (JSON/Pydantic)
- Integrate AI with Amazon Connect (contact flows, IVR, agent assist, administration) and related services (Lex, Transcribe, Comprehend where applicable)
- Deploy scalable, cost-efficient solutions using Lambda and Fargate; front with API Gateway; automate with CloudFormation
- Establish observability, testing, and evaluation: tracing, prompt/version management, offline/online LLM eval, A/B testing
- Enforce security and compliance best practices: IAM least privilege, KMS encryption, VPC endpoints/PrivateLink, data redaction and PII handling
- Collaborate with product and CX teams to translate requirements into reliable, measurable AI-enabled features
- Document architectures, runbooks, and playbooks; provide knowledge transfer to operations teams
Required Skills and Experience:
- Advanced: Agentic frameworks and Generative AI (prompt engineering, tool use, structured outputs, guardrails)
- Advanced: Python for LLM orchestration; experience with LangChain, Semantic Kernel, or similar
- Intermediate: Retrieval-Augmented Generation (RAG) design and LLM evaluation (faithfulness, relevance, toxicity, latency/cost)
- Strong: REST API development with FastAPI; async patterns; unit/integration testing
- AWS services: CloudFormation (IaC), API Gateway, Lambda, Fargate, OpenSearch
- AWS Bedrock for chatbot/Q&A use cases, model selection, prompt templates, grounding strategies
- Amazon Connect administration and contact flow design; integrating AI for caller self-service and agent assist
- CI/CD with Git-based workflows; logging/monitoring with CloudWatch, X-Ray; container build and packaging (Docker)
Preferred Qualifications:
- Experience with embeddings (Bedrock Titan, SageMaker, or open-source), vector schema design, hybrid search
- Familiarity with Amazon Lex, Kendra, S3 data lakes, DynamoDB, Step Functions, EventBridge
- Production-grade evaluation pipelines and feedback loops (golden sets, human-in-the-loop, telemetry-driven improvement)
- Knowledge of data governance, security controls, and responsible AI practices
- Performance tuning and cost optimization for serverless and container workloads
- Contact center analytics and quality automation leveraging Connect data
Education and Experience:
- 4+ years building backend or data-intensive applications on AWS, including 1–2+ years with LLM/GenAI solutions
- BS in Computer Science or related field, or equivalent experience
Success Indicators:
- Measurable improvements in answer quality and containment rate for Q&A and contact center scenarios
- Low-latency, cost-efficient APIs with high reliability and clear observability
- Secure, compliant deployments with reusable templates and automated pipelines
Work Arrangement:
- Flexible; remote or hybrid depending on location and team needs
Never miss an opportunity! Create an alert based on the job you applied for.
