Frontier AI Engineering

Sourced from postman.wd108.myworkdayjobs.com

Postman is building a small Frontier AI Engineering team to work directly with ambitious customers on bringing real AI and agentic systems into production. This hands-on role spans the full AI stack—from models and inference to agents, APIs, tools, context, evaluation, security, and production infrastructure. Engineers prototype, debug, and ship alongside customer teams while feeding recurring patterns back into Postman's AI platform and product roadmap.

• Build production AI systems using LLMs, agents, MCP, APIs, tool calling, RAG, vector search, and multimodal models
• Design agentic systems with controls around identity, permissions, policy, and execution
• Experiment across foundation models, model routing, inference, prompt/context engineering, and fine-tuning
• Build evaluation systems for agents and LLM applications covering task success, tool correctness, hallucination detection, safety, latency, and cost
• Make AI production-ready with observability, tracing, guardrails, fallbacks, rate limits, retries, caching, and model gateways
• Solve retrieval, memory, grounding, embeddings, and context window challenges for customers
• Secure agent actions via authentication, authorization, secrets, policy enforcement, and auditability
• Own problems end to end from architecture to prototype to evaluation to production
• Turn recurring customer patterns into reusable primitives and platform capabilities

• Deeply technical, hands-on builder comfortable writing code and debugging agent traces
• Strong engineering fundamentals and AI fluency across the AI stack
• Experience with LLMs, agents, RAG, APIs, and production AI workflows
• Background in applied AI, agentic systems, AI infrastructure, distributed systems, developer platforms, APIs, security, ML infrastructure, inference, or model platforms
• Comfort working with ambiguity and taking problems from undefined to production
• Curiosity, judgment, speed, and ownership

• Experience with MCP-based integrations
• Experience building evaluation frameworks for production agents
• Experience with model routing across multiple LLM providers
• Experience with identity and policy layers for agent actions