Summary: Seeking a Senior AI/ML Backend Developer to build and scale production-grade AI agent systems using AgentCore and Strands, designing robust RAG pipelines and integrating AWS Bedrock models. Highlights: 1. Build and scale production-grade AI agent systems 2. Design robust RAG pipelines and high throughput data processing services 3. Integrate AWS Bedrock foundation models in low latency, reliable APIs **Title: Senior AI/ML Backend Developer** **Position Type: Contract/ Full Time** **Location: Remote across Canada** **About the Role** We're looking for a Senior AI/ML Backend Developer to build and scale production grade AI agent systems using AgentCore runtime and the Strands agent framework. You'll design robust retrieval augmented generation (RAG) pipelines, high throughput data processing services, and integrate AWS Bedrock foundation models in low latency, reliable APIs. **What You'll Do** * Architect, build, and optimize AI agents using AgentCore runtime and Strands for task orchestration, tool usage, and multi step reasoning. * Design and deploy RAG pipelines (chunking, embeddings, vector stores, ranking) with strong observability and evaluation. * Implement scalable Python microservices and data processing (ETL/ELT) with async I/O, queuing, and backpressure. * Integrate AWS Bedrock models (e.g., Anthropic/Meta/Amazon) with safe guardrails, prompt orchestration, and cost controls. * Productionize agents with CI/CD, IaC (Terraform/CDK), Amazon EKS or ECS/Fargate, and automated testing. * Establish evaluation \& governance for agents/RAG (accuracy, latency, hallucination, drift, PII handling). * Collaborate with Product, Data, and Security to hit performance SLAs and compliance standards. **What You'll Bring** * 6 10\+ years backend engineering; 3\+ years AI/ML apps/agents in production. * Hands on with AgentCore runtime and Strands agent framework (or equivalent agent frameworks). * Deep RAG experience (vector DBs like Pinecone, OpenSearch/KNN, Milvus, or pgvector). * Strong Python (FastAPI, asyncio, typing, pydantic), API design, and testing (pytest). * Cloud native on AWS (EKS/ECS, Lambda, SQS/Kinesis, Step Functions); AWS Bedrock familiarity. * Data processing at scale (Spark/Flink or Dask preferred), and feature/log pipelines. * Security \& compliance mindset (PII handling, key management, IAM, audit trails).