Frontier AI Engineer

Postman, Inc. is building a Frontier AI Engineering team to help customers bring AI and agentic systems into production. The role involves hands-on engineering with customers and turning recurring AI engineering challenges into reusable Postman product capabilities.

Responsibilities:

• Build real AI systems. Work with LLMs, agents, MCP, APIs, tool calling, RAG, vector search, structured outputs, multimodal models, and production AI workflows
• Design agentic systems. Build agents that can reason across tools, APIs, enterprise data, and workflows while maintaining the right controls around identity, permissions, policy, and execution
• Work across the model layer. Experiment with leading foundation models, model routing, inference, prompt and context engineering, fine tuning where it makes sense, and the tradeoffs between quality, latency, cost, and reliability
• Make AI measurable. Build evaluation systems for agents and LLM applications, including task success, tool correctness, hallucination detection, regression testing, safety, latency, and cost
• Make AI production ready. Work on observability, tracing, guardrails, fallback strategies, rate limits, retries, caching, model gateways, and the infrastructure needed to operate AI systems reliably at scale
• Solve the context problem. Help customers reason about retrieval, memory, grounding, context windows, embeddings, vector databases, enterprise data, and how the right context gets to the right model or agent at the right time
• Secure agent actions. Work on authentication, authorization, secrets, API access, policy enforcement, human approval, auditability, and the controls required when AI systems can actually take actions
• Own problems end to end. Start with a problem that may not be clearly defined and take it from architecture to prototype to evaluation to production
• Stay close to the code. We are looking for builders. You will prototype, integrate, debug, experiment, benchmark, and ship
• Work with ambitious engineering teams. Spend time with AI teams, platform teams, developers, and engineering leaders who are pushing the boundaries of what they can build
• Shape what Postman builds next. The problems you see in the field will directly influence our AI platform, APIs, Gateway, identity, agent capabilities, developer experience, and product roadmap
• Turn learning into a product. Identify patterns that repeat across customers and turn them into reusable primitives, reference architectures, workflows, and platform capabilities
• Help define Frontier AI Engineering at Postman. We are early. The first engineers who join will help define the technical bar, architecture, engineering practices, operating model, and culture of the team

Requirements:

• We are looking for people who are deeply technical and love building
• People who are just as comfortable opening a terminal and debugging an agent trace as they are sitting with another engineering team and working through an architecture
• People who understand that building with LLMs is not just about writing prompts. It is about systems engineering, evaluation, reliability, data, APIs, security, and everything required to make probabilistic systems behave predictably in production
• People who are curious about where models are going, but equally interested in what it takes to make them useful today
• People who like hard problems and do not need everything perfectly defined before they start building
• Your background might be in applied AI, agentic systems, AI infrastructure, distributed systems, developer platforms, APIs, security, ML infrastructure, inference, or model platforms
• We care about strong engineering fundamentals, AI fluency, technical depth, curiosity, judgment, speed, and ownership