Artificial Intelligence Engineer

About the role

Own and evolve a production conversational AI system using a LangGraph tool-calling agent, guardrails, and RAG at scale. Requires 3+ years of production backend experience including a shipped LLM feature, async typed Python, agent framework experience, PostgreSQL and pgvector, live-service operations, tracing, retries, and advanced testing/evaluation. Nice to have Azure, Terraform, vector search at scale, guardrail frameworks, React/TypeScript, and data privacy engineering.

Requirements

• 3+ years building production backend services
• At least one shipped LLM-backed feature
• Strong async Python with mypy strict typing discipline
• Hands-on LangChain, LangGraph, or similar agent framework experience
• Solid PostgreSQL schema design, performance tuning, migrations, and connection pooling
• Experience operating live services and diagnosing issues from telemetry
• AI security boundaries and server-side authorization
• Debugging with traces and correlation IDs
• Retries, backoff, and circuit breakers
• Contract tests and evaluation of nondeterministic systems

What you'll do

• Extend conversational AI pipeline, guardrails, query reformulation, embedding-based routing, and agent orchestration
• Maintain internal API tool layer with schemas, authorization, pagination, and error handling
• Support production system, triage incidents, investigate latency and quality regressions, improve telemetry and runbooks
• Improve RAG with PostgreSQL full-text search and pgvector
• Build model and prompt evaluation, adversarial datasets, shadow traffic, canary rollouts, and rollback
• Reduce LLM cost and latency with telemetry, prompt trimming, caching, and model tiering
• Strengthen tenant data security and prompt-injection defenses
• Expand automated testing across release gates

Where this was posted

• LinkedIn

  last seen 1h ago
  View on LinkedIn