We’re building a platform of autonomous AI agents for data collection, labeling, and intelligent analysis — and we’re looking for a Senior Backend Engineer to power this system. You will architect and implement core services that interface with LangGraph-based workflows, manage data access from structured and unstructured sources, and support seamless orchestration between our AI layers and data infrastructure. The ideal candidate is comfortable working with AWS Bedrock, vector databases, and building APIs optimized for AI-scale interaction.

Responsibilities

• Build performant, production-grade backend APIs and services using AWS Lambda, EC2, and RDS.
• Integrate structured and unstructured data access layers with LangGraph and LangSmith agents.
• Architect retrieval pipelines using vector databases to enable semantic search for AI agents.
• Coordinate with data engineers to support RAG workflows across a shared data lake.
• Develop and maintain task routing, memory management, and agent orchestration logic.
• Contribute to platform-level observability, metrics, logging, and diagnostics for agentic tasks.
• Deploy and manage MCP servers to wrap scraping, proxy, and JS-rendering tooling as standardized Model Context Protocol (MCP) services.

Required Skills

• 4+ years of backend development experience, with strong programming skills (Python, Node.js).
• Deep experience with serverless architectures, microservices, and AWS (Lambda, EC2, RDS).
• Familiarity with LangGraph, LangChain, or equivalent AI orchestration frameworks.
• Working knowledge of vector databases (Chroma, Pinecone) and semantic search.
• Solid database and API design expertise, particularly with PostgreSQL.
• Experience building scalable, low-latency backend systems for AI or data platforms.

Preferred Qualifications

• Experience with agentic AI systems or orchestration of multi-agent workflows.
• Understanding of RAG patterns in production.
• Comfortable with Databricks and Unity Catalog data integration strategies.
• Exposure to AWS Bedrock and large model orchestration in cloud-native environments.