AI/ML Intern

Build practical AI features — chatbots, retrieval (RAG), document automation — on top of models like Claude and GPT, and ship them into real products rather than notebooks.

• Prototype and ship LLM features: prompts, tool use, structured output, evaluation
• Build retrieval pipelines (chunking, embeddings, pgvector) over real documents
• Measure quality: build small eval sets, compare prompts and models, track cost and latency
• Integrate AI features into Django/Next.js apps with proper error handling and fallbacks
• Document what works and what doesn't, so the team can reuse it

• A paid monthly stipend
• Real work on live products (alokknight.com, SmartLandlord) and on client projects — not practice tasks
• Direct mentorship from the founder, a senior engineer who built and runs a SaaS live in 7 European markets
• Internship certificate and a letter of recommendation on successful completion
• A full-time offer for the strongest performers
• Remote and flexible: 3–6 months, full-time preferred, part-time (20+ hours/week) possible for final-year students

• Apply with your CV and the links asked for below
• A short practical task (a few hours, in your own time)
• A 30-minute video call with the founder
• Offer and start date — usually within a week of the call

• Strong Python; comfortable with HTTP APIs and JSON
• Practical understanding of how LLMs work: context windows, tokens, prompting, embeddings
• A small AI project of your own we can look at — GitHub or demo link
• Honest about limitations — we ship things that work, not demos
• Available 3–6 months, at least 20 hours a week

• Experience with the Anthropic/OpenAI APIs, LangChain or LlamaIndex
• Vector databases
• FastAPI or Django
• Basic ML with scikit-learn or PyTorch