Haoci Zhang

Software Engineer

Seattle, Washington, United States7 yrs 11 mos experience
Highly Stable

Key Highlights

  • Led development of machine learning infrastructure at Meta.
  • Expert in optimizing training frameworks for recommender systems.
  • Built scalable tape storage systems for archival needs.
Stackforce AI infers this person is a Machine Learning Engineer with a strong focus on SaaS and optimization technologies.

Contact

Skills

Core Skills

Machine LearningSoftware Engineering

Other Skills

Parallelism optimizationsHeterogeneous trainingSemi-sync trainingAttention mechanismsOptimizersSpeculative decodingFlashAttention kernelsSparse embedding lookup optimizationsReal-time MLTTFBTape storageArchival storageResearchC/C++Javascript

Experience

7 yrs 11 mos
Total Experience
2 yrs 6 mos
Average Tenure
3 mos
Current Experience

Stealth ai startup

Software Engineer

Mar 2026Present · 3 mos · Bellevue, WA

Meta

Senior Staff Software Engineer

Feb 2019Mar 2026 · 7 yrs 1 mo · Bellevue, WA

  • 2024–2026: Pretraining Infra@MSL, also TL@PyTorch Training Enablement team; productionized numerous optimizations in the parallelism space such as zero-bubble pipeline parallelism and fwd–bwd overlapping, supported team on explorations on heterogeneous and semi-sync training, and drove productionization of novel attention, optimizers, speculative decoding and many other architectural changes.
  • 2021–2023: TL on Recsys training infra for multiple internal customers; developed custom FlashAttention kernels for Recsys, productionized new training frameworks for Ads and delivered core optimizations such as sparse embedding lookup optimizations, real-time ML, TTFB etc.
  • 2019–2020: Built tape storage from 0→1 and also supported HDD fleet within Archival Storage, productionizing tape-based storage and large-scale restore, repair, rebalance, and recovery pipelines.
Parallelism optimizationsHeterogeneous trainingSemi-sync trainingAttention mechanismsOptimizersSpeculative decoding+8

Facebook

Software Engineering Intern

May 2018Aug 2018 · 3 mos · Seattle, Washington

  • Didn't do anything useful :)

Columbia university in the city of new york

Research Intern

Jun 2016Oct 2016 · 4 mos · New York, New York

Walmart ecommerce

Software Engineering Intern

Jul 2015Sep 2015 · 2 mos · Sunnyvale, California

  • Didn't do anything useful :)

Education

Columbia University

Master of Science - MS

Jan 2017Jan 2018

Tsinghua University

Bachelor’s Degree — Computer Science

Jan 2013Jan 2017

Stackforce found 100+ more professionals with Machine Learning & Software Engineering

Explore similar profiles based on matching skills and experience