John Hartquist

• •Rust implementation of Alexandre François's Resonate algorithm: per-bin spectral analysis without FFTs or windowing.
• •Cross-platform: native on macOS, Linux, iOS, plus browser (WASM). Python (PyO3) and JavaScript bindings.
• •Outperforms reference implementation (C++, Apple-only) by 1.6x on M2 Max while remaining portable.
• •Live WebGL streaming-spectrogram demo runs on desktop and mobile.
• •Write-up: johnhartquist.com/resonators.

• •Converts rustdoc JSON into markdown so AI coding agents can query crate docs without spending tokens parsing HTML.
• •Works for local projects, external dependencies, and any crates on crates.io.
• •Ships an installable skill for coding agents.

• •Runs tasks autonomously in isolated git worktrees with verification (tests, clippy, build) gating merges.
• •Retry-with-learnings: failed attempts produce structured findings that feed forward into the next attempt's context.
• •267 tests, clippy-pedantic clean.

• •End-to-end Rust ports of CREPE, Swift-F0, and Spotify's Basic Pitch via Hugging Face's Candle framework.
• •Numerical-parity tested against the original TensorFlow / PyTorch implementations.

• •Cal Poly MS thesis (2012): real-time polyphonic guitar transcription using Non-Negative Matrix Factorization.
• •Full Rust rewrite, GuitarSet evaluation, neural experiments with HCQT features and Mamba-family architectures.
• •Adapted Spotify's Basic Pitch to run in real-time in the browser via `resonators` front-end.

Remote

• •Shipped features across TakeOne's ML video pipeline: real-time on-device tracking, video segmentation (SAM, Track-Anything, GroundingDINO), color correction and matting, audio cleanup (DeepFilterNet + FFmpeg loudness normalization), transcription (Whisper, Deepgram), and LLM-driven script editing.
• •Worked across the stack: Python ML, Swift iOS, Kotlin Android, Django backend, Next.js user-facing applications, and AWS GPU inference (EC2, ECS, Batch).
• •Built a transcript-editing pipeline with an LLM-as-judge: structured Pydantic rubrics, hallucination-catching validation gates, and retry-with-feedback up to 5 attempts.
• •Wound the company down in March 2026.

Kirkland, WA

• •Prototyped AR measurement and virtual-object simulation in Unity for in-store and at-home use cases.
• •Built React / Babylon.js inspector and ingestion tool for 3D home assets; published a custom open-source product viewer.
• •Built a room-scanning prototype using Apple's RoomPlan API to capture 3D room geometry; researched data stores (EdgeDB, Solid) for structured home-asset modeling.

Seattle, WA

• •Productionized a financial-reconciliation ML model (Python, XGBoost) and migrated the pipeline to Spark / Scala for >100× speedups over the prior Python implementation.
• •Built tools for managing datasets, training models, and evaluating performance, enabling researchers to iterate rapidly on reconciliation accuracy.
• •Built a full-stack platform for running Jupyter notebooks in production (TypeScript, React, Next.js, Docker, GCP).

San Francisco, CA

• •Took a year off to study classic ML and deep learning in depth (FastAI parts 1–2, Andrew Ng's Coursera specializations).
• •Built `fastai_audio` (194★), a library for on-the-fly GPU spectrogram generation during training; write-up.

San Francisco, CA

• •Built licensing and brand-protection platform (Ruby on Rails, Angular) for HDMI, Dolby, and Philips.
• •Engineered data model, optimized SQL queries, and developed real-time dashboards in a multi-tenant architecture.

Mountain View, CA

• •Built iOS app (Objective-C) and Python/Django backend services for an early-stage 'Tinder for Jobs' startup.

San Luis Obispo, CA

• •Built internal services for the CreateSpace Royalties & Financials team (Java, Spring, AWS) using test-driven development.