SDE Intern
BharatCode
IndiaFresher6m agovia LinkedIn
About the Role
BharatCode is seeking an SDE Intern to work on optimizing LLM inference on limited hardware.
Key Responsibilities
- Working on the inference stack: model serving, batching, scheduling, and caching.
- Enhancing response speed and maximizing GPU utilization.
- Contributing to the BharatCode desktop app, CLI, and API.
- Developing real-time server monitoring tools.
Requirements & Tech Stack
- Strong programming skills in Python.
- Comfortable reading code in other languages.
- Familiarity with Linux, Git, and the command line.
- Interest in how LLMs operate on hardware.
- At least one demonstrable personal project.
- Full-time availability for a 3-month internship, remote.
Qualifications
- Currently pursuing or recently completed a degree in CS, IT, or a related field, or equivalent hands-on experience.
Good to have
- Experience with serving frameworks like vLLM, SGLang, or TensorRT-LLM.
- Understanding of GPUs, CUDA, batching, KV caching, or quantisation.
- Experience with Docker, monitoring tools, or production services.
- Open source contributions or past hackathon projects.
What you get
- Stipend: ₹60,000/month for 3 months.
- Ownership of infrastructure used by thousands of students.
- Direct work with the founding team.
- Internship certificate and recommendation letter for strong performers.