Nuvama's Quant Engineering team builds internal platforms that power quantitative research and systematic trading. We are expanding our investment in AI-driven tooling and are looking for engineering interns who can contribute across the full stack — from production APIs and React interfaces to LLM integration and fine-tuning pipelines.

Duration: 6 months

What You'll Work On

Full stack product development: building and iterating on internal research and analytics platforms using React, TypeScript, and Python (FastAPI). Includes dashboards, data visualisation components, real-time data feeds, and workflow UIs.
LLM integration and agent development: integrating large language models into internal tooling via API (function-calling, multi-turn conversations, tool use). Building agent backends that connect LLMs to internal data sources and services.
Fine-tuning pipelines: constructing supervised fine-tuning (SFT) datasets from internal corpora and running fine-tuning jobs using LoRA/QLoRA on open-source base models (Llama 3, Mistral, Phi-3). Evaluating outputs on domain-specific benchmarks.
Retrieval-Augmented Generation (RAG): building and evaluating RAG pipelines over internal document and data stores using vector search. Experimenting with chunking strategies, embedding models, and retrieval quality metrics.
Backend services and APIs: building REST and WebSocket APIs, async task workers, and data ingestion pipelines that serve the research and trading platforms.
Infrastructure and tooling: contributing to deployment automation, monitoring dashboards, and developer environment improvements.

What We're Looking For

Pursuing B.Tech / M.Tech in CS, AI/ML, or a related field from a Tier 1 institution.
Strong Python fundamentals. Comfortable building REST APIs and working with async code.
Working knowledge of React — hooks, state management, component patterns. TypeScript is a plus.
Genuine hands-on interest in LLMs: you should have experimented with prompt engineering, RAG, or fine-tuning in a personal or academic project.
Conceptual understanding of transformer architecture — attention, tokenisation, context windows. You should be able to reason about model behaviour, not just call APIs.
Comfortable working on Linux servers, reading logs, and debugging distributed services.

Nice to Have

Hands-on experience fine-tuning open-source LLMs (Llama, Mistral, Phi) using PEFT/LoRA, Axolotl, or similar frameworks.
Experience building agent systems with function-calling or multi-step reasoning.
Familiarity with vector databases and RAG evaluation frameworks.
Prior projects involving AI-powered internal tools, research assistants, or automated workflows.
Exposure to managed LLM inference platforms (Azure AI Foundry, AWS Bedrock, or similar).

Full Stack & AI Engineering Intern