Full Stack AI Engineer

What you’ll do

• Build and own full-stack AI features across frontend, backend, and data layers for web applications.
• Design agentic workflows (single- and multi-agent / A2A) that can plan, route, call tools, and coordinate to complete complex tasks.
• Implement and refine RAG pipelines, including retrieval strategies, chunking, embeddings, reranking, and hybrid search across multiple data sources.
• Design and operate GraphRAG-style retrieval on top of knowledge graphs to support multi-hop reasoning and relationship-heavy use cases.
• Build NL2SQL / NL2DB capabilities that convert natural language into safe, validated queries against SQL databases, warehouses, or analytics systems.
• Define and manage tool interfaces and MCP-style capability layers so agents can call internal APIs, SaaS tools, and data services with proper contracts and permissions.
• Create evaluation pipelines for prompts, agents, RAG, GraphRAG, NL2SQL, and tool use, including regression tests, LLM-as-judge scoring, and human review loops.
• Instrument AI systems with traces, logs, metrics, and structured events so you can debug failures, track versions, and understand behavior across the entire request path.
• Build dashboards and alerts to monitor quality, latency, cost, and safety signals for AI features in production.
• Collaborate with product, design, data, and platform teams to move from prototype to production while adding guardrails, fallbacks, and human-in-the-loop flows where needed.
• Continuously experiment with new models, prompting techniques, and architectures, then distill what works into reusable patterns and libraries for the team.

Required qualifications

• Hands-on experience shipping LLM-based features (agents, RAG, tool calling, or NL2SQL) into production.
• Strong full-stack engineering experience with modern web stacks (e.g., TypeScript/React/Next.js plus Python/Node.js).
• Solid backend fundamentals: REST/GraphQL APIs, relational databases, caching, and cloud deployment (AWS/GCP/Azure with Docker/CI).
• Experience designing, measuring, and improving AI behavior using evals, metrics, and real user feedback.
• Ability to work closely with product teams, own projects end to end, and make pragmatic tradeoffs between quality, speed, and cost.