Jimmy Yao

Co-founder, Voluma Bio
Previously AI/ML at Epic Systems

About

I’m a software engineer focused on applying AI and LLMs to the real world.

Right now I’m co-founding Voluma Bio, where we’re building agents to take over the computational analysis after a lab experiment: segmenting the scans, placing the landmarks, running the measurements, with a reviewer agent checking each step.

Before that I spent three years at Epic Systems shipping LLM summary workflows for clinical chart review across 100+ hospital systems.

Away from the keyboard: tennis, the great outdoors when I can get out into it, and the firm position that Roronoa Zoro is the GOAT.

Work

Voluma Bio

Co-founder · Jul 2026 – Present

Building AI agents that automate the computational analysis after lab experiments.

  • Built a TypeScript multi-agent runtime where agents run multi-turn tasks and execute domain code in isolated Daytona/Docker sandboxes.
  • Prototyped LabRat, which automates a mouse-knee micro-CT workflow (3D bone segmentation, anatomical landmarks, osteoarthritis measurements) while an independent reviewer agent gates each phase.

Epic Systems

Software Developer, AI/ML · Aug 2023 – Jul 2026

Shipped LLM summary pipelines for clinical chart review across 100+ hospital systems, summarizing 300 million notes in one month.

  • Led a referral-focused LLM workflow from clinician interviews through context assembly, prompt design, orchestration, and review UX.
  • Built a synthetic-data evaluation framework that generates targeted test cases for extraction quality, summarization accuracy, prompt-injection resistance, and bias coverage.
  • Developed a blinded multi-LLM review system to triage pipeline outputs, flag evaluation failures, and route ambiguous cases to clinicians.
  • 68% of surveyed clinicians said the summaries highlighted patient information they otherwise would have missed.

M&T Bank

Software Engineer Intern · Jun 2022 – Dec 2022

Built an AWS-deployed UX interview analysis pipeline (Transcribe, BERT/T5 summarization, sentiment, topic clustering) with a React dashboard for reviewing results.

AIH, LLC

Software Engineer · Jan 2020 – Jun 2022

Owned the health-data backend (Flask, Docker, MongoDB, AWS), built the aiSpine posture-wearable Android app, and ported speech and facial stroke-screening models to a Flutter app with TensorFlow Lite.

University of Rochester

2019 – 2023

BS Computer Science; Minor in Mathematics and Business

Projects

Meridian CLIGitHub stars: 30

Feb 2026 – now · Python, TypeScript, MCP

Launches and tracks specialist agents across Claude Code, Codex, OpenCode, Cursor, and Pi, with session forking, recursive delegation, and file-based memory for long-running agents.

Mars Agents

Mar 2026 – now · Rust

A package manager for agents, skills, and hooks: resolves a lockfile, compiles harness-native configuration, and never deletes files it did not create.

creative-writing-skillsGitHub stars: 536

2026 · Claude Code skills

A Mars package of agents and skills for long-form fiction: planning, drafting, critique, and continuity.

Meridian Flow

Oct 2025 – now · TypeScript, React, Yjs, PostgreSQL

An AI writing platform where each agent thread is a Yjs CRDT peer, editing live documents alongside humans or a review draft the writer applies atomically.

Meridian Reader

Sep 2026 – now · Swift, ONNX Runtime, Core ML

An offline iOS read-aloud app running Phonon TTS on Core ML and Kokoro on ONNX Runtime. Moving Kokoro off Core ML cut its crashes from 5 in 20 calls to 0 in 320, at 3.3× real time on an iPhone 17 Pro.

Tether

Oct 2026 – now · Rust, TypeScript, Swift

A phone-first tmux client. A Rust daemon fans tmux control-mode output to web and iOS clients with per-connection backpressure and snapshot reconnect.

Research

Temporal QA agent benchmark

Research

Benchmarked BM25, hybrid retrieval, and tool-using Gemma 4 agents on TemporalWiki. Agents reached 95% temporal recall@5 but about 20% exact match: evidence use, not retrieval, was the bottleneck.

LPS sensitizes breast cancer response to IAP antagonist therapy

Frontiers in Immunology, 2022

Co-first author. Contributed wet-lab experiments, data analysis, and manuscript writing.

LlamaIndex EPUB parser

Open source

Added .epub ingestion to LlamaIndex/LlamaHub using EbookLib and html2text.

Writing

75% of LLM Reads Are Redundant

May 12, 2026 · draft

I scraped 3 months of AI agent sessions across Claude Code, Codex, and OpenCode. 75% of file reads were redundant.

A Working-Memory Cache for Agent File Reads

May 12, 2026 · draft

A proposal: let the harness hold one current copy of each active file, instead of storing every read in the conversation.

What Happened to My Tokens?

May 11, 2026 · draft

198 agent spawns ran overnight and half my Codex plan was gone by morning. The culprit was a parent agent that couldn't just wait.

All notes

Jimmy Yao builds AI agents for lab science at Voluma Bio. Previously shipped LLM workflows for clinical chart review at Epic Systems.

haowjy.github.io