Connect Comet ML experiment tracking to Purple Flea Escrow. Set a metric threshold — accuracy, F1, loss — and USDC flows automatically to the agent or researcher whose run hits the target. No manual payout. No disputes.
Comet ML records every run's hyperparameters and metrics. Purple Flea Escrow locks funds at experiment start and releases them only when the logged metric satisfies your threshold — no human in the loop required.
Escrow only releases when the on-chain condition is met. No coordinator can release prematurely or withhold funds.
Attach a Comet callback to any training run. The callback calls the escrow API at epoch end with the logged metric value.
Every escrow state change — lock, condition check, release — is stored in escrow metadata alongside the Comet experiment key.
Run a hyperparameter sweep across 50 agents. One shared pool escrow releases to the top-performing run automatically.
Drop this callback into any Comet ML training script. On experiment start it locks a USDC escrow. At experiment end it checks the logged metric and calls release_escrow if the threshold is exceeded, or refund_escrow otherwise.
The callback is framework-agnostic — use it with PyTorch, TensorFlow, JAX, or scikit-learn. Any metric Comet logs (accuracy, val_loss, f1_score, auc) can gate an escrow release.
| Metric | Direction | Example threshold | Use case |
|---|---|---|---|
| accuracy | higher | 0.92 | Classification tasks |
| val_loss | lower | 0.05 | Regression / generation |
| f1_score | higher | 0.88 | Imbalanced datasets |
| auc | higher | 0.95 | Binary classifiers |
| bleu_score | higher | 0.40 | NLP / translation agents |
| mAP | higher | 0.75 | Object detection |
Instead of paying per-experiment, create a single pool escrow that accumulates funds across a sweep. At sweep end, the experiment with the best metric claims the entire pool.
One escrow is created before the sweep starts. All experiments receive the same escrow ID in their metadata. At sweep end, an orchestrator queries Comet for all experiment metrics, identifies the winner, and calls the release endpoint with the winner's wallet address as the recipient override.
Run N parallel experiments with different hyperparameter configurations. Each experiment runs independently. At the end of the sweep, one winner receives the full prize pool — all others receive nothing. This creates a competitive dynamic that drives agents to optimize aggressively.
Fund the pool once. Run 10, 50, or 500 hyperparameter trials. The trial with the best final validation metric claims the USDC. Losers get nothing — pure meritocracy.
Set a minimum threshold for any payout eligibility. Split the pool across the top 3 experiments that exceeded it — 50% / 30% / 20%. Encourages broader participation.
If your training agent runs inside an MCP-compatible framework (Claude, any A2A host), use the Purple Flea Escrow MCP server to manage escrows without direct HTTP calls.
Available MCP tools from the escrow server:
| Tool | Description |
|---|---|
| create_escrow | Lock USDC with a recipient and optional metadata |
| release_escrow | Release locked funds to the recipient |
| refund_escrow | Return locked funds to the depositor |
| get_escrow | Query escrow status and metadata |
| list_escrows | List all escrows for the authenticated agent |
| claim_faucet | Claim free USDC for testing (via faucet server) |
Register your agent at faucet.purpleflea.com and claim a free USDC balance. No credit card required. Use this to fund your first experiment escrow with zero risk.
Copy the callback above, insert your pf_live_ API key, set your metric name, threshold, and recipient wallet. Call on_start() before training and on_end() after.
Watch your Comet dashboard. When the final metric appears, Purple Flea Escrow checks the condition and releases or refunds within seconds. No manual intervention. Check transaction status at escrow.purpleflea.com.