AI Inference Framework Engineer

About the role

Quadric is redefining edge AI with the industry's first General Purpose Neural Processing Unit (GPNPU), enabling developers to run both neural network inference and conventional C++ code on a single programmable architecture. Our technology powers intelligent edge devices across automotive, industrial, robotics, and embedded systems.

Founded in 2016 and based in downtown Burlingame, California, Quadric is building the world's first supercomputer designed for the real-time needs of edge devices. Quadric aims to empower developers in every industry with superpowers to create tomorrow's technology, today. The company was co-founded by technologists from MIT and Carnegie Mellon, who were previously the technical co-founders of the Bitcoin computing company 21.

THE OPPORTUNITY

Quadric's GPNPU runs both neural networks and C++ code on a single programmable architecture. As a member of our AI Application team, you will make the GPNPU a first-class backend in the frameworks developers already use, such as llama.cpp, PyTorch, and vLLM. That way, customers can move their applications from GPUs to Quadric with minimal changes. You'll own backend integration from memory and dispatch through serving features, working closely with our kernel, runtime and compiler teams.

WHAT YOU'LL DO

Integrate GPNPU backend into AI compute frameworks (llama.cpp, PyTorch, vLLM, Optimum)

- Implement framework backend interface, memory allocation, compute dispatch, registration and operators/kernels

- Implement model compute pipelines including prefill, decode, batching, multi-sequence and multi-modal

- Support important framework serving features including KV-cache management, paged attention, continuous batching, speculative decoding

Optimize framework performance and accuracy for GPNPU backend

- Reduce host-device dispatch overhead, implement graph capture/replay, integrate efficient kernels, minimize latency

- Support quantization formats such as GGUF (Q4/Q8), GPTQ/AWQ, int8/fp16

- Validate accuracy against reference backends (perplexity, task benchmarks) and drive improvements

WHAT SUCCESS LOOKS LIKE

Within your first 6-12 months, you’ll:

- Bring up a new LLM/VLM family end to end on the GPNPU backend, with accuracy matching the reference

- Deliver measurable tokens/s or latency gains on a frontier model

- Ship one serving feature (e.g. paged KV cache) in a production release

WHAT WE'RE LOOKING FOR

REQUIRED

- 3+ years building or extending ML frameworks or inference runtimes

- BS/MS in CS/CE or related fields

- Strong C/C++ and Python

- Solid understanding of framework internals: tensor memory management, graph scheduling, operator dispatch

- Solid understanding of AI compute performance analysis and optimization

- Fluent with AI coding agents (e.g. Claude Code, Cursor), and interested in building agent-driven workflows for backend and model bring-up Preferred