Applied AI Intern

About Vel Labs

Vel Labs is a small team building personal intelligence: technology that remembers, understands and grows with the people it serves. Our first product, Sola, is a companion that learns how you think and helps you live more intentionally.

About the role

We’re building at the intersection of AI and consumer technology, and we’re looking for an Applied AI Intern to design, build and ship LLM-powered systems to production with the rigour of a full-time engineer.

You’ll work with the founders and the core AI team on a conversational product people use every day across text, voice notes, images and real-time calls. This role goes beyond prototypes: you’ll own systems that run on live traffic and answer for their quality, reliability and latency.

What you’ll do

• Design and improve multi-step LLM pipelines: context assembly, planning and structured prompts
• Build long-term memory and retrieval, and decide what context reaches the model on each turn
• Implement agent tools with clear contracts, timeouts, retries and failure behaviour
• Work on multimodal and real-time voice systems where latency, streaming and interruptions matter
• Define evals, build eval sets and use traces and production data to fix quality regressions

What we’re looking for

• Strong TypeScript or Python, and comfort working in both
• Hands-on experience building on LLM APIs: structured outputs, tool calling and streaming
• A solid grasp of async programming: concurrency, cancellation, timeouts and error handling
• Working knowledge of REST APIs, SQL and Git, and a habit of testing for edge cases
• Shipped work we can review: projects, open-source contributions or prior internships
• Nice to have Agents, retrieval or memory systems, LLM evals, or speech and vision models.

Why join us

• High ownership Own production AI systems that directly shape the user experience.
• Founder access Work closely with the founders on technical and product decisions.
• End-to-end exposure Work across model integration, system design, evals, deployment and monitoring.
• Engineering rigour Work in a codebase held to high standards for testing, review and observability.
• Growth Exceptional performance can lead to a full-time pre-placement offer.

How to apply

If you have a strong engineering foundation, a serious interest in applied AI and the discipline to take systems all the way to production, we’d love to hear from you. Use the application form, or email careers@vellabs.in with a link to your work and a short note on why you’d be good at this role.