🤫 × Thinking Machines · the plan

What we start today.

A plan we can begin the moment this publishes — and a clear line from here to build together, sell together, and buy together. Trust first, always. Nobody worth working with buys from a stranger.

Start a conversation The alignment

A forward-looking proposal from public information — not a claimed deal, commitment, or endorsement.

The order we run

Trust → pilot → build → sell & buy.

00Open the conversation, stand up the bench

Now · 0–30 days

Publish this dossier so our team is aligned on Thinking Machines' science, products, and worldview from day 0.
Stand up a provider-neutral integration bench inside 🤫 Agent One over MCP: Claude Code, OpenAI Codex, Grok 4.5, and Gemini, side by side.
Prototype "teach-your-own-agent": fine-tune an open-weight model (Tinker-style) on a 🤫 Puppy One developer unit — private, on owned compute.

01A small, real pilot

1–3 months

One concrete, consented use case where a person teaches their own private model on compute they own.
Make it reproducible and receipted end to end — deterministic inference, every access logged, nothing leaves without consent.
Earn a real relationship first — no ask before trust.

02Build together

3–9 months

Bring the science of efficient, reliable fine-tuning into the 🤫 owned-compute stack, on open rails (the Hushh Protocol / PCHP).
Experiment with real-time interaction models for the always-on, voice-first 🤫 Agent One.
Keep publishing in the open, in the shared spirit of Connectionism.

03Sell together, buy together

9+ months

Joint go-to-market on open, inspectable rails — never a walled garden.
Buy together on compute where it lowers cost per watt and per workload for everyone.
Grow the 🤫 Factory One grid so more people own the means of their own intelligence.

The ask

One conversation.

We are not asking for a deal. We are asking for a conversation — thirty minutes between Mira Murati and Manish Sainani to compare notes on a future we describe almost identically and build from two complementary ends. The science of teaching your own model, meeting the place that model should live: owned, private, at the edge.

Everything above is a proposal we can start today, on our side, whether or not that meeting happens. But it would be a great deal more fun to build it together.

Start a conversation Read the essay

Sources & the fine print

Public facts, cited.

This is a public study and an open, honest invitation from 🤫 hushh — an admiration of Thinking Machines' public work and a proposal to explore a partnership. It is not a claimed deal, affiliation, endorsement, or sponsorship. Every fact here is drawn from Thinking Machines' own public writing or reputable press; figures that are press-reported are labeled "reported", and quotations are verbatim from the cited source. Companies and facts change — verify against the primary sources before acting.