Data Engineer - Oracle to PostgreSQL Re-Platform

Stack

Responsibilities

• ▹Own end-to-end delivery: take a production LLM agent from requirement to production deploy, defending architecture and trade-offs with the client
• ▹Build and harden agent orchestration (LangGraph/LangChain or equivalent) — routing, tool-calling, planning, synthesis, state management
• ▹Integrate tools over MCP and maintain a growing tool surface, including BM25, hybrid, or vector retrieval
• ▹Run models on AWS Bedrock and Bedrock AgentCore — model selection/routing, guardrails, memory, regional residency
• ▹Build the evaluation harness (golden sets, LLM-as-judge, CI quality gates) and instrument the system with OpenTelemetry for token, cost, and latency attribution
• ▹Drive down cost and latency through model routing, prompt caching, payload pruning, and parallelizing independent calls
• ▹Design and stage safe write-actions with least-privilege permissions, human-in-the-loop approval, plan versioning, audit trail, and rollback

Requirements

• ▹5+ years in software engineering, at least 2 at a senior level
• ▹Strong Python in production, comfort with TypeScript or Go when needed
• ▹Hands-on LLM application engineering: prompt design, tool/function calling, structured output, context management, token budgeting
• ▹Agent orchestration experience with LangGraph, LangChain, or hand-rolled orchestration
• ▹Managed LLM/agent platform experience in production: AWS Bedrock, Google Vertex AI, or Azure AI Foundry
• ▹Evaluation, retrieval & observability experience: eval harnesses, golden/reference sets, vector or hybrid search in production, OpenTelemetry-based distributed tracing
• ▹Production AWS, CI/CD & IaC experience, including Terraform in production
• ▹Working English, comfortable defending system design on client calls

Soft skills