About the Role
We're looking for an AI Developer Intern to help build applied-AI features into real products. You'll work with Large Language Models (LLMs) across multiple providers — designing prompts, generating structured output, building tool-calling/agentic workflows, and wiring it all into live applications. This is practical AI engineering: turning LLM capabilities into reliable, production-ready product features, with a focus on quality, cost, and safety.
Key Responsibilities
- Build and improve AI-powered features — content generation, rewriting, summarization, structured data extraction, and assistant/chat experiences.
- Design, test, and iterate on prompts (system + user), and maintain versioned/reusable prompt systems.
- Implement structured output pipelines — LLM → JSON → schema validation (Zod/Pydantic) → retry-on-failure — so AI output is reliable enough to use in production.
- Work with multiple LLM providers (OpenAI, Anthropic, Gemini, OpenRouter, Ollama, or any OpenAI-compatible endpoint) through a provider-agnostic abstraction, including tool/function calling.
- Help with evaluation and quality — measuring output quality, latency, token usage, cost, and fallbacks.
- Integrate AI features into applications via API routes in TypeScript/Python, alongside the web team.
- Implement guardrails and safety (content constraints, safe URL/data fetching, graceful fallbacks).
Required Skills
- Strong Python and/or TypeScript/JavaScript programming fundamentals.
- Hands-on experience calling LLM APIs (OpenAI, Anthropic, Gemini, Ollama, or similar) — including from personal projects.
- Solid grasp of prompt engineering and the difference between chat, structured/JSON output, and tool/function calling.
- Understanding of JSON, schemas, and data validation (Zod, Pydantic, or equivalent).
- Basic understanding of RAG, embeddings, tokens, temperature, and context windows.
- Comfort reading and writing code in a real application (not just notebooks); Git workflow.
- Nice to have: Vercel AI SDK, LangChain/LlamaIndex, vector databases (pgvector/Qdrant), running local models (Ollama), evals/tracing, or fine-tuning experience.
Qualifications
- Pursuing or recently completed a degree in Computer Science, AI/ML, Data Science, or a related field.
- Projects, GitHub, notebooks, or coursework demonstrating applied AI / LLM work (chatbots, RAG apps, generation pipelines, etc.).
- Able to work on-site in Nikol, Ahmedabad, full-time for the internship duration.
- Curiosity about applied AI product-building and comfort experimenting toward measurable quality.
Salary & Benefits
- Monthly stipend: ₹10,000–₹18,000 (based on skill and interview performance) (placeholder — confirm/adjust to your budget; AI roles often command a slight premium).
- Direct experience building production LLM features on a modern AI stack, senior mentorship, and a Certificate of Internship.
- Pre-placement offer (PPO) / full-time conversion for high performers.
Pay: ₹2,000.00 per month
Work Location: In person
