AI, blockchain, and the Go underneath both.

Products

Where a model actually runs

Every figure in these demos came out of a real model — nothing mocked, nothing stubbed. Some call one while you watch; in others the model ran before you arrived and the page shows what it produced. In three, the model's ruling lands on Base Sepolia, verbatim and permanent.

In the escrow, a seller releases funds with a Groth16 proof of a delivery secret — the secret is never revealed, not even to the contract. When delivery is contested, the case goes to an AI arbiter: it reads the evidence both parties published on-chain, rules, and its full reasoning is written back to the chain verbatim, permanently.

The arbiter's power is bounded by the bytecode. It can credit a pull-payment balance and nothing else — it cannot push funds, cannot touch an escrow that isn't disputed, and cannot revisit one that's settled. That's the part worth looking at: not that an AI made a decision, but that the decision is auditable and the authority is enforced by a contract rather than a policy.

No screenshots, no mock data. Contract addresses are on BaseScan, source is on GitHub, and every settlement is sitting there to read.

• ZK Escrow Explorer

  Groth16 release, AI-arbitrated disputes.

• AI Parametric Insurance

  Language models adjudicating a claim, checkably.

• Deposit Dispute

  Models rule on evidence; the contract does the arithmetic.

• Bench Lab

  A benchmark that says where it was wrong.

• Judge Lab

  An eval that mostly measured its own harness.

• SBA Lab

  A model guessed which industries default. Then the federal data graded it.

• Access Audit Lab

  An access audit you re-run in the browser, filed under real controls.

• Agent Gateway

  A policy gate for AI agents. Edit it, and diff what moved.

• RAG Lab

  A RAG pipeline, measured — failures included.

• LangDAG Lab

  A conversation with a branch in it is a tree, not a list.

• Slack AI Lab

  Watch a Slack bot answer, from outside the workspace.

Live on Base Sepolia

Contracts deployed and readable right now — every figure below is fetched from the network on load, and every address resolves on BaseScan. No model is involved in any of them.

• RWA Tokenization Explorer

  ERC-3643-style KYC-gated transfers.

• ZK KYC Pass Explorer

  Prove a credential without revealing an identity.

• Gas Optimization Race

  Same contract, twice, raced live on-chain.

• Interest Rate Model Explorer

  An immutable lending curve, read live.

• Oracle Health Explorer

  Live Chainlink feeds, and what is wrong with them.

• MMR Log

  A commitment that never grows, measured.

• Cosmos Lab

  Cosmos proofs, replayed and refused.

See how these contracts are tested →

Go & systems labs

Runnable labs for the awkward parts of systems work — backpressure, profiling, reconciliation, scheduling, and the Rust ownership rules that trip everyone up. Each one is driven in the browser and computed from real code.

• unselect2

  Replaces the <select> but looks like and behaves like the <select>

• WASM Keygrind

  Go compiled to WebAssembly, grinding keys in the tab.

• Backpressure Lab

  Three answers to a full queue, raced live.

• Aegean Placement

  A proven bound next to an algorithm without one.

• pprof Lab

  What a block profile sees that a CPU profile cannot.

• Reconcile Lab

  What a reconcile loop does when a write is refused.

• Orbit Lab

  Orbital mechanics at frame rate, in the browser.

• Kafka Lab

  Why a rate limit cannot fix a slow consumer.

• Goroutine Visualizer

  Press kill on a goroutine and watch nothing happen.

• Rust Lifetime Lab

  Where rustc stops, and where Go moves the value to the heap.

• Rust Move Lab

  Who owns this value? Ask rustc, then watch the Go program run.

• Rust Borrow Lab

  Two refusals, and the program the borrow checker accepts.

• Rust Send/Sync Lab

  A guarantee on one side, a detector that watches on the other.

• ETL Lab

  State that is never kept is state that cannot go stale.

• AMM Lab

  The curve, the fee, and what a sandwich bot takes.

• Arb Lab

  Two pools drift apart. How much of the gap is actually yours?

• Agent Swarm Lab

  One controller or a mesh: which swarm gets through?

• CRDT Lab

  Edit offline on three replicas. Do they still agree?

• Wallet risk triage & sanctions screening

  Screen an address. Triage a wallet. Draft the escalation.

• P2P Lab

  Two browsers, one relay. Fetch a block by its hash.

• WebSocket Lab

  One upstream, many subscribers. Measured, not estimated.

• FIX Lab

  Your own FIX session. Lose a message, watch the engines recover it.

• RPC Gateway

  One endpoint, several providers. Knock one out and watch the next answer.

Practical tools

Online tools built and run in-house — from OSINT and security checks to SEO and marketing analyzers. Most need no account; the ones with a limit say so on the page.

• TraceCheck

  Trace an email header end to end and see every hop it really took.

• Domain Recon

  Pull DNS, WHOIS, and mail records for any domain in one pass.

• EXIF Metadata

  Reveal the camera, location, and edit history hidden inside an image.

• Fake Recruiter Check

  Test whether a recruiter message is a real opening or a scam.

Not another AI wrapper.

Every tool here hits real infrastructure — DNS, WHOIS, TLS, raw metadata — before any model reasons over it. Evidence first, interpretation second, and you can verify every signal yourself. Same rule on-chain: the arbiter's authority is whatever the contract grants it, and its reasoning is public record.

Why Pigfox is different →

What Pigfox delivers

Engagements start with an outcome rather than a stack: an AI integration that earns its place, a smart contract you can audit before you trust it, or the Go backend underneath them. Pigfox designs it, builds it, ships it to production, and stays on it afterwards.

• AI engineering & AI-augmented development
• Smart contracts, tested and fuzzed before they ship
• Go/Golang APIs & backend systems
• APIs & Proxies
• Web sites & applications
• Rebuilding or upgrading outdated business systems
• WASM applications
• Organic Search Engine Optimization
• Practical tools — OSINT and security checks, SEO and marketing analyzers

Build something together