AI Engineer, Internship - Summer 2026 - Applications Open Now

Who Are We? Postman is the world’s leading API platform, used by more than 45 million+ developers and 500,000 organizations, including 98% of the Fortune 500. Postman is helping developers and professionals across the globe build the API-first world by simplifying each step of the API lifecycle and streamlining collaboration—enabling users to create better APIs, faster. The company is headquartered in San Francisco and has offices in Boston, New York, Austin, Tokyo, London, and Bangalore - where Postman was founded. Postman is privately held, with funding from Battery Ventures, BOND, Coatue, CRV, Insight Partners, and Nexus Venture Partners. Learn more at postman.com or connect with Postman on X via @getpostman. P.S: We highly recommend reading The "API-First World" graphic novel to understand the bigger picture and our vision at Postman. The Opportunity (Summer 2026 AI Internship - Applications Open Now) We're seeking an  AI Engineer Intern to work alongside our AI team on large-scale AI and Agentic  systems from data pipeline to production deployment. This role is scoped for someone with foundational experience who wants to deepen it: you'll own discrete pieces of real systems under the mentorship of senior engineers, not shadow work or isolated coursework-style projects. What You'll Do You’ll work directly with the AI team, taking responsibility for well-scoped pieces of real systems, with mentorship from senior engineers. Benchmarks & Evaluation Contribute to APIFlow-Bench , our open-source benchmark for real API-development work: design and review benchmark tasks and their mock API environments, extend the evaluation harness and task-generation pipeline in Python, and help maintain the public multi-model leaderboard with statistical confidence intervals. Help build a new action-level AI safety benchmark: instead of grading what a model says, it scores what an agent actually does inside a simulated enterprise API environment. You’ll work on scenario design, threat modeling (prompt injection, data exfiltration, permission overreach), and auditable evaluation design. Model Training & Efficiency Fine-tune open-weight models for tool calling and agentic tasks (SFT, distillation, and RL) using PyTorch and the open-source training ecosystem, on both managed training platforms and self-managed cloud GPUs. Design and run experiments with rigor: evaluate every training run on our benchmarks, support ablation studies and error analysis, track experiments, and report results honestly, including cost. Evaluate ultra-low-bit quantized models for on-device use: extend our quantized vs. full-precision benchmark comparisons and analyze where and why they diverge. Agent Systems & Engineering Practice Help build the next generation of Postman’s in-product AI agent (Agent Mode): a deliberately minimal agent architecture that calls LLM APIs directly (tool loops, multi-step execution, checkpointing), primarily in TypeScript. No prior TypeScript is required; strong Python fundamentals transfer quickly. Read the source code of open-source agent harnesses and turn what you learn into design specs and prototypes. Document experiments, design decisions, and runbooks so your work is legible to the next person; flag safety, fairness, or privacy concerns you observe in model or agent behavior. About You Currently pursuing a BS, MS, or PhD in Computer Science, Data Science, or a related quantitative field. Hands-on experience training or evaluating ML models: course projects, research, hackathons, or a prior internship all count. Solid Python fundamentals: data structures, functions, basic testing; comfortable writing and reviewing code outside of notebooks. Working knowledge of at least one deep-learning framework (PyTorch preferred). Clear written and verbal communication, and a habit of documenting what you build. Preferred Qualifications   (none required; the more of these you have, the better) Experience fine-tuning open-weight LLMs (SFT, LoRA, RL, or distillation), with the improvement measured on a benchmark. Experience building LLM agents (tool calling, multi-step loops) or LLM evaluation harnesses/benchmarks, and reporting results with statistical rigor. A track record of shipping real software end-to-end: APIs and services, CLIs, Docker, CI/CD, cloud; public code on GitHub is a big plus. Interest or experience in AI safety and robustness: red-teaming, prompt injection, agent security, fairness, or interpretability. Exposure to model-efficiency work: quantization, low-bit inference, or serving optimization. Evidence of rigor and initiative: publications, technical blog posts, ablation studies, or self-driven side projects with quantified results. Fluency with AI coding tools (Claude Code, Cursor, Codex) to ship fast while still deeply understanding the systems you build.   What Else? In addition to Postman's pay-on-performance philosophy, and a f