Zack Li logoZack Li

Zack Li

@zhiyuan8

Entrepreneur, builder, lifelong learner — CTO & Co-founder of Nexa AI (acquired by Qualcomm).

I'm Zack Li. I build scalable AI systems end to end — from model training and software architecture to Gen AI inference that ships on device.

I'm a Senior Staff Machine Learning Engineer at Qualcomm, leading GenieX, the generative AI inference runtime that runs frontier models across Snapdragon NPU, GPU, and CPU. I also contribute to llama.cpp and support new GenAI models on Qualcomm® AI Hub.

I cofounded Nexa AI as CTO, building Nexa SDK (now GenieX) into a leading on-device AI runtime—#1 on GitHub Trending, #1 Product of the Day on Product Hunt, and 8k+ GitHub stars. I co-authored the Octopus and Omni model series—top-2 trending on Hugging Face and showcased at Google I/O and by Google DeepMind. I drove enterprise adoption with customers like HP and Lenovo, and led engineering delivery alongside partners including Qualcomm, AMD, NVIDIA, Microsoft, and more.

Before Nexa, I built on-device AI and software systems at Google, Amazon Lab126, and Cadence. I hold an M.S. from Stanford University (2019) and a B.E. from Tongji University (2017).

Zack Li

Selected work

AI Infra

Nexa SDK (GenieX)

A GenAI inference runtime that runs any frontier model on the edge in one line of code, across NPU, GPU, and CPU. Hit #1 on GitHub Trending and #1 Product of the Day. Our app Hyperlink, built on it, reached 2.1M views and 30K users in two months. Recognized by Qualcomm, AMD, NVIDIA, Intel, IBM, and Microsoft.

NexaSDK / GenieX SDK architecture diagram
Model Training

On-device agentic AI models: a pioneering series using functional tokens for strong function calling across phones, PCs, wearables, cars, and robots. Reached Hugging Face's top two trending models, and spotlighted at Google I/O and Google DeepMind.

Blog posts

Selected publications

Industrial recognition

Chipmakers
Qualcomm (Blog) recognition screenshot
Qualcomm (Blog)

Developer blogs on Snapdragon Android, the Hexagon NPU, Granite 4.0, and IoT & robotics.

Qualcomm (LinkedIn) recognition screenshot
Qualcomm (LinkedIn)

Featured on Qualcomm's X and LinkedIn, plus seven "This Week in AI" posts (1, 2, 3, 4, 5, 6, 7).

NVIDIA recognition screenshot
NVIDIA

RTX blogs on our local agent and CES booth, a 1.3M-view video.

AMD (Blog) recognition screenshot
AMD (Blog)

Blogs on NPU image generation and DeepSeek R1 speedups with NexaQuant.

AMD (LinkedIn) recognition screenshot
AMD (LinkedIn)

Posts from AMD, AMD Enterprise, and AMD Developer, plus an interview on its developer channel.

Intel recognition screenshot
Intel

Spotlighted NexaSDK on its developer channel for on-device LLMs on Intel's NPU.

Model labs
Google DeepMind recognition screenshot
Google DeepMind

Our model OmniAudio is featured by Google DeepMind in its Gemmaverse, a showcase of models built on Gemma.

Google (Blog) recognition screenshot
Google (Blog)

Highlighted in the 2024 Gemma keynote, with posts from Google AI Developers and Google Devs.

OpenAI recognition screenshot
OpenAI

Its Developer Relations team amplified our work running open models on consumer hardware.

Hugging Face recognition screenshot
Hugging Face

Ranked our models among the most-downloaded from 2022 to today; CEO Clément Delangue called to collaborate.

IBM recognition screenshot
IBM

Listed NexaSDK as a supported framework for Granite 4.0, next to vLLM, llama.cpp, and MLX.

Alibaba recognition screenshot
Alibaba

The Qwen team called out our day-0 Qwen3-VL support and Qwen3 on the Qualcomm NPU.

Liquid AI recognition screenshot
Liquid AI

Partnered with us on the LFM2.5 launch and LFM2.5 Thinking, amplified by a Qualcomm VP.

pyannoteAI recognition screenshot
pyannoteAI

Ran their diarization model on Qualcomm NPUs through Nexa SDK for CES 2026.

Operating systems
Microsoft recognition screenshot
Microsoft

Demoed our product on stage at the Ignite 2025 keynote and listed us as a software partner for Windows.

Apple recognition screenshot
Apple

The Apple MLX team reposted our work running multimodal models on Apple silicon.

Industry Talks

Experience

Senior Staff Machine Learning Engineer · QualcommJoined through Qualcomm's acquisition of Nexa AI.
2026 — Present
CTO & Co-founder · Nexa AI
2023 — 2026
Software Engineer III · Google
2022 — 2023
Software Engineer II · Amazon Lab126
2021 — 2022
R&D Engineer · Cadence Design Systems
2019 — 2021
Teaching Assistant, CS229T Machine Learning Theory · Stanford University
2018

© Copyright 2026 Zack Li.