Senior AWS Agentic AI Engineer – Snowflake

About the Role

We’re looking for a Senior AWS Agentic AI Engineer to design, build, deploy, and operate enterprise-scale AI agents and multi-agent systems on AWS. You’ll help build a scalable agentic AI platform that connects autonomous agents to enterprise tools, APIs, data platforms, and business workflows. Strong Snowflake and Snowflake Cortex knowledge is a plus, since agents will need secure access to enterprise data.

We want an AWS-first agentic AI engineer, not just an LLM developer: someone who can take an agent from architecture through development, deployment, security, tool integration, evaluation, and production, while building reusable patterns that can support hundreds of agents and thousands of users.

What You’ll Do

• Design and build production-grade AI agents and multi-agent systems on AWS, covering reasoning, planning, tool calling, memory, workflow execution, and human-in-the-loop controls
• Implement supervisor, worker, router, planner, and specialist agent patterns, including orchestration, delegation, state management, and agent-to-agent communication
• Build event-driven, asynchronous agent architectures and reusable frameworks, tools, prompts, and agent templates
• Optimize agents for latency, scalability, reliability, token usage, and cost
• Build secure agent runtimes with Amazon Bedrock and Bedrock AgentCore (Runtime, Gateway, Memory, Identity, observability), plus Bedrock Guardrails and Knowledge Bases
• Decide when an agent should run on Lambda, EKS, or AgentCore based on workload, latency, state, and scale
• Design and develop MCP servers and clients, including secure tool discovery, RBAC, least-privilege access, and enterprise MCP gateway and registry patterns
• Integrate agents with AWS services, databases, Snowflake, SaaS platforms, and enterprise APIs
• Design enterprise RAG architectures using S3, OpenSearch, Bedrock Knowledge Bases, and vector databases, with hybrid search, metadata filtering, reranking, and retrieval evaluation
• Implement AI security and governance: agent identity, secrets management, encryption, network isolation, PII/PHI protection, prompt-injection and jailbreak defenses, tool authorization, and audit logging
• Build end-to-end observability across agent runs, LLM and tool calls, token consumption, latency (P95/P99), errors, cost, and task success
• Build AI/agent CI/CD pipelines (code, prompt, model, tools, security, evaluation, deployment, observability) with regression testing, LLM evaluation, and blue/green or canary releases
• Securely connect agents to Snowflake while preserving RBAC, governance, lineage, and auditability

Required Qualifications

• 7+ years of software engineering experience
• 3+ years of hands-on GenAI/AI engineering
• Strong Python skills; strong SQL
• Strong AWS cloud engineering experience (Lambda, EKS, API Gateway, EventBridge, SQS/SNS, Step Functions, S3, OpenSearch, DynamoDB/RDS, IAM, Secrets Manager, KMS, CloudWatch, CloudTrail, VPC)
• Hands-on Amazon Bedrock experience
• Experience building production AI agents
• Strong understanding of LLMs and agent architectures, including prompt engineering, tool/function calling, structured outputs, context management, embeddings, RAG, agent memory, and model routing
• Experience with RAG and vector search
• Strong API, microservices, and distributed systems experience
• Strong AWS IAM and security knowledge
• Experience with CI/CD and DevOps (GitHub/GitLab, Docker, Infrastructure as Code)

Preferred Qualifications

• Bedrock AgentCore experience
• MCP development
• LangGraph, LangChain, or Strands Agents (ideally framework-agnostic design experience)
• EKS and Kubernetes; Terraform or AWS CDK
• Event-driven architecture and multi-agent systems
• OpenTelemetry, LangSmith, or Fiddler
• Snowflake: Cortex (Analyst, Search, Agents), Snowpark, semantic models, AI functions, vector search
• Experience with Anthropic Claude, Amazon Nova, or Meta Llama models on Bedrock
• TypeScript/JavaScript, Java, or Bash

Pay: $60.00 - $70.00 per hour

Expected hours: 40.0 per week

Work Location: Remote