YC Summer 2026 (S26) founder

Niki Kotecha

Founder of Lamb Labs, Lightning fast chips with hardcoded AI models

Company

Niki Kotecha is listed as a founder of Lamb Labs. Lamb Labs is building MPUs (Model Processing Units), custom chips that hardcode LLM weights directly into the chip. Unlike GPUs, which move weights between memory and compute, MPUs keep the model on-chip, targeting up to 20,000+ tok/s and a 63Γ— higher intelligence per watt.

Founder traction evidence

Public posts and activity attributed directly to this founder.

  1. X

    custom chips at 20,000 + tok/s!!! building the future of hardware for ai inference πŸ‘πŸ‘

    custom chips at 20,000 + tok/s!!! building the future of hardware for ai inference πŸ‘πŸ‘

  2. X

    spot the lambs outside of the HQ, on a road trip to meet customers πŸ‘πŸ‘

    spot the lambs outside of the HQ, on a road trip to meet customers πŸ‘πŸ‘

  3. X

    Researcher ➑️ Venture Scientist Look forward to starting my journey at @conceptionxtech in Cohort 6.

    Researcher ➑️ Venture Scientist Look forward to starting my journey at @conceptionxtech in Cohort 6.

  4. X

    yc female founder brunch, apparently highest % of female founders in the batch πŸš€

    yc female founder brunch, apparently highest % of female founders in the batch πŸš€

  5. X

    24 days left till demo day @ycombinator happy sunday working from the @LambLabs HQ followed by a founder dinner!

    24 days left till demo day @ycombinator happy sunday working from the @LambLabs HQ followed by a founder dinner!

  6. X

    gap is closing between open weights and closed models...

    gap is closing between open weights and closed models... feels like the moat is slowly diminishing...

  7. X

    finally learnt the difference between whilst and while

    finally learnt the difference between whilst and while

  8. X

    @bscholl from @boomsupersonic at @ycombinator earlier this week.

    @bscholl from @boomsupersonic at @ycombinator earlier this week. biggest takeaway: always be scrappy & keep maximising your surface area of serendipity πŸ‘