# MAYA CHEN
# Staff Infrastructure Engineer · Brooklyn, NY
# github.com/maya-chen  ·  maya@chen.dev  ·  maya.chen.dev

## SUMMARY
Staff-level infrastructure engineer. 7y of open-source contribution.
Author/co-author of 3 repos · 6,843+ stars · 184 collaborators.

## CORE LANGUAGES
  Rust         ██████████ 31.2%
  Python       ████████ 24.8%
  TypeScript   ███████ 21.4%
  Go           ████ 13.0%
  Shell        ██ 4.6%

## NOTABLE WORK

  [01] ferrium/task-queue
       role:    Author & maintainer
       lang:    Rust
       stars:   2,421  ·  forks: 187  ·  contribs: 32
       desc:    Distributed task queue with at-least-once delivery, in Rust. Powers async work for ~40 backend services.
       tags:    distributed-systems, rust, async, tokio

  [02] lumen/ml-pipelines
       role:    Core maintainer
       lang:    Python
       stars:   1,184  ·  forks: 96  ·  contribs: 28
       desc:    Pipeline toolkit for shipping ML models to production — feature stores, model registries, eval harness.
       tags:    mlops, python, kubernetes

  [03] edge-fn/runtime
       role:    Co-author
       lang:    TypeScript
       stars:   1,893  ·  forks: 142  ·  contribs: 19
       desc:    WASM-based edge function runtime with sub-ms cold starts. Sandboxes untrusted workloads on shared infra.
       tags:    wasm, edge, v8, isolates

  [04] grok-replay
       role:    Author
       lang:    Go
       stars:   487  ·  forks: 31  ·  contribs: 9
       desc:    Record/replay debugger for gRPC services. Captures real prod traffic and replays against canaries.
       tags:    grpc, go, debug

## CAREER TIMELINE
  2021   Joined the `lumen` org
         Started maintaining ml-pipelines. Wrote the feature store layer that 7 other teams now depend on internally.
  2022   Co-authored `edge-fn/runtime`
         Two-engineer side project that grew into a startup's core product. Sub-ms cold starts via WASM + V8 isolates.
  2023   `grok-replay` + first conference talk
         Built a record/replay debugger for gRPC. Presented at StrangeLoop — first time speaking publicly about her work.
  2024   LLM eval infra at scale
         Pivoted into LLM platform work. Built `eval-harness` and wired it into CI for 6 production LLM apps.
  2025   Current focus
         Reducing edge runtime cold-start from 0.8ms → 0.2ms. Mentoring two new maintainers on `task-queue`.

## ACTIVITY METRICS
  commits:          12,384
  PRs merged:          487
  repos:                64
  longest streak:       61 days

# generated by gitbio · grounded in 12,384 commits