Echo ProgramLog in

The pilot

Your models, our foundation. Eight weeks, at no cost, on training jobs you would otherwise curate blind.

Before the compute is spent, one call tells you whether curating toward your target will pay. Timestamped before you train, checked against what your run does.

How it works
  1. 01

    Send two samples

    A seeded random sample of whole documents from your target and from your candidate pool, a few million words in all. Our reference client draws them and records the seed and the full corpus size. The corpus never leaves your machines.

  2. 02

    Receive the verdict

    In seconds, on CPU, without touching your model. One of three calls, a confidence tier, a scope card, and a forecast id. On Select, a scoring table comes back so your pool is ranked on your own hardware in a single pass.

  3. 03

    Train, then tell us

    Where you can, train both arms: the curated sample and a random sample of the same size. Report the held-out result under the forecast id. The receipt ties the call to the outcome, and the receipt is the product.

The three calls
Select
Curating toward this target is forecast to help.
Curate. Take the scoring table and keep the top of your pool.
Do not select
Curating is forecast to backfire, or the target carries no usable signal.
Train on a broad random sample of the same size and keep the compute.
No forecast
Too close to the calibrated boundary to call.
Treat the gain as a coin flip. Add target data and run again.

Each call carries a confidence tier, the scope card, and a fingerprint of the sample it was made on. Every forecast is timestamped before your first training step.

The exchange
What you get
  • A key on our hosted service. Nothing to install, nothing of yours to deploy.
  • Fifty calls a day, raised on request. Samples travel compressed, far more room than a forecast needs.
  • Verdicts with scoring tables, local keep lists through the reference client, and outcome receipts.
  • A written forecast-versus-outcome report at the end, for every run you report.
  • A direct line to the founder for interpretation and for fitting the call into your pipeline.
What we ask
  • Run the forecast on at least three real target and pool pairs.
  • Where you can, train both arms and report the held-out result. A forecast without an outcome teaches neither of us anything.
  • Tell us what a verdict must say to be acted on: formats, sizes, latency, wording.
Data
What travels
  • Samples only, over HTTPS, processed in memory and never written to disk.
  • The target needs about 2,600 words in two or more documents. The pool sample at least four times that. More target data sharpens the reading.
What we keep
  • Document and word counts, a fingerprint of the sample so a receipt is tied to exactly the text it was made on, the verdict, timestamps, the sampling seed, and any outcome you report.
  • Never your text. Never the scoring table. Keys are stored hashed and revoked on request, one per organisation.
Validated range
Text models
up to 41.7M parameters, from scratch
Image and text
one 256M-parameter model, continued training
Data kept
a quarter of the pool
Language
English, whitespace-delimited text

Every response carries this card. Tell the client the model size, keep fraction and language you plan to use; when your job is outside the range, the confidence drops one tier and the response says why. The call itself does not change. Larger models are exactly what the pilot is for.

Terms
Apply

A few lines are enough. We read every application and answer by hand.

Contact

Questions before applying, a scope you are not sure fits, or a pipeline that needs the call in a different shape: write to us. A person answers.

[email protected]

Members reach the founder through the same address, and every email from the program can be replied to.