Generative AI Intern

Role overview Build and test language model features as part of real engineering projects, with support from a senior engineer and a defined project scope. The work ranges from rapid prototyping to evaluating model behavior and helping investigate issues in deployed agent systems.

Responsibilities - Prototype Python features for retrieval, information extraction, and agents that use tools. - Create evaluation datasets, scoring scripts, and regression checks, then use them to compare prompts and models. - Contribute to retrieval pipelines, including document chunking, embeddings, and ranking. - Record experiments and results clearly so others can reproduce them. - Work with engineers to investigate unsuccessful agent runs using production traces.

Requirements - Studying or recently graduated in computer science, data science, AI, or a related field, or equivalent practical experience. - Solid Python skills and comfort with Git and the command line. - Some hands-on experience using language model APIs, including through personal or academic projects. - A careful approach to testing and checking whether outputs are correct. - Clear written communication.

Nice to have - A demo or code repository showing a language model project. - Experience with vector databases, FastAPI, or workflow frameworks. - Familiarity with evaluation methods such as test sets, scoring rubrics, or model-based judging.

Benefits and work setup - Remote internship lasting 3–6 months. - Pairing with a senior engineer and exposure to how production agent systems are built, tested, and monitored.