AI / LLM Engineer Intern
Build with large language models — prompts, agents, and evals that ship to real users.
About the role
We're teaching software to test itself, and LLMs are at the heart of it. As an AI/LLM Engineer Intern you'll build real features on top of models like Claude and Gemini — prompts, agents, tool use, and the evals that keep them honest.
You'll work in our TypeScript codebase with senior engineers, ship code that real customers run, and learn the parts of applied AI that don't show up in tutorials: reliability, guardrails, and measuring whether a prompt actually works.
It's a 3-month paid internship based in Lahore (Gulberg III), hybrid — three days a week onsite, the rest remote. Strong interns have a clear path to a full-time engineering role.
What you’ll do
- Prototype features on top of LLMs (Claude, Gemini) with the team
- Write and refine prompts, then measure them with evals
- Help build small agent workflows and tool integrations
- Explore retrieval, structured outputs, and safety guardrails
- Read, test, and improve real production code in TypeScript
What we’re looking for
- Studying CS/SE or a related field — or a strong self-taught builder
- Solid programming fundamentals (Python or JavaScript/TypeScript)
- Genuine curiosity about LLMs and how to make them reliable
- You read documentation and learn fast
Nice to have
- Side projects using the Anthropic / OpenAI / Gemini APIs
- Familiarity with Git, REST APIs, or Node.js
- A grasp of evals, embeddings, or vector search
Apply for this role
Tell us a little about you and attach your CV — it takes two minutes.