Skip to content
View tishachawla-jg's full-sized avatar
🎯
Focusing
🎯
Focusing
23 followers · 17 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: StarstruckAchievement: YOLO

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: StarstruckAchievement: YOLO

Block or report tishachawla-jg

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
tishachawla-jg/README.md

Tisha Chawla

Co-Founder AgentPlane, Ex-Microsoft  ·  building the control plane for AI agents
https://tishachawla.com
Co-creator of TokenOps and Chronicle at The Agent Plane

The Agent Plane LinkedIn dev.to YouTube

The problem I work on

Multi-agent systems fail in two ways that ordinary tooling cannot see. They overspend, because a runaway loop burns a month of budget before a per-request rate limit notices anything is wrong. And they are unreproducible, because a trace tells you what happened without letting you run it again.

We built the two pieces of infrastructure that fix that, as open source.

🪙 TokenOps

Run-aware token governance. Cap spend across a whole agent workflow instead of per request, with a shared ledger plus in-path enforcement that can throttle, downgrade, or halt a run before it burns the budget.

stars PyPI

pip install agent-tokenops

🎬 Chronicle

Record and replay for agent decision graphs. Turn a production failure into a committed regression test, then re-run your fix deterministically, with no live LLM calls and no flaky reproduction.

stars PyPI

pip install agent-chronicle

Featured by Microsoft Developer: "Who spent all the tokens?"   LinkedIn · X

Both feed a shared control plane, one SQLite and one UI, where agents never touch the database file. Releases are exercised against testbench, a set of tool-neutral multi-agent workloads I keep separate so the testing stays honest.

🎤 Talks
FinOps for AI Agents: Who Spent All the Tokens? AI Engineer World's Fair, San Francisco · with Susheem Koul
Your Agent Failed in Prod. Good Luck Reproducing It. On replayability for autonomous agents · with Susheem Koul
Shorts → Short-form agent infrastructure explainers

✍️ Writing

Working on agent infrastructure, FinOps for LLMs, or reproducible agent testing? I'd like to hear about it. Say hi on LinkedIn.

Pinned Loading

  1. theagentplane/chronicle theagentplane/chronicle Public

    Record-and-replay for agent decision graphs: reproduce a prod agent failure as a committed regression test, and re-run your fix without live LLM calls.

    Python 27 6

  2. theagentplane/tokenops-wiki theagentplane/tokenops-wiki Public

    Python 4

  3. SEC-EDGAR_Analysis_App SEC-EDGAR_Analysis_App Public

    App Demo -

    Python 1

  4. theagentplane/tokenops theagentplane/tokenops Public

    Run-aware token governance for multi-agent systems.

    Python 84 23