In development
Abodra
Building-design software for exploring a property project and preparing a builder handoff.
In development Last verified 2026-10-18.
Featured projects
Products I am developing and engineering tools you can inspect. Each card includes its status and available links. Client results are separate.
Product · Scan verification pending
ShipReady
A launch-check tool for AI-built apps. It separates findings from declared information and checks that still need a person.
Product · Scan verification pending Last verified 2026-10-18.
Beta candidate
AssetForge 3D
A workspace for creating and refining 3D assets from prompts and reference images.
Beta candidate Last verified 2026-10-18.
Open-source engineering project
CipherClaw
An open-source toolkit for investigating recorded multi-agent runs.
RepoOpen-source engineering project Last verified 2026-10-18.
More engineering work
Open source · MIT
ClawKeeper
An agent governance project that checks access, permissions, and approvals before a model runs. The public repository shows the rules and audit-log design.
Live demo and public repo. Last verified 2026-08-07.
Related TechTide engagement: Systems Decision Audit at TechTide AI.
Open source · Apache 2.0
TechTide Harness Kit
Schemas and test fixtures for defining what an AI agent may do. Covers tool permissions, network access, human approval, and prompt-injection tests.
Live docs and public repo. Last verified 2026-08-07.
Related TechTide engagement: Systems Decision Audit at TechTide AI.
Open source · Apache 2.0
Swarm 357
357-agent orchestrator governed by per-agent budget caps, a pattern-matched bash security gate, auto-downgrade daily budgets, a hard per-run spend ceiling, and correlation-ID telemetry.
Live site and public repo. Last verified 2026-08-07.
Private build
AgentRewind
Time-travel debugging for agent traces: zero-trust import parsing with signature verification and size clamps, five trace formats auto-detected, and a divergence/latency/token diff engine.
Private build. Technical writeup published. Last verified 2026-08-07.
Open source
FintheFinder
Deep-research agent with a real human-in-the-loop approval workflow, where the run suspends until a human approves, plus a deterministic citation audit with zero AI in the verification path.
Public repo. No hosted demo. Last verified 2026-08-07.
Private build
Molten
Governed agent runtime: kill switch with cost and error auto-triggers, circuit breakers, a four-tier execution sandbox, hard daily and per-task budget caps, and a dead-man's switch.
Private build. Technical writeup published. Last verified 2026-08-07.
Related TechTide engagement: Systems Decision Audit at TechTide AI.
Open source
TechTideAI2
Company-scale agent platform on a three-plane architecture covering control, data, and evidence. State-machine status transitions, hard execution timeouts, adversarial eval harness with a 33-task golden suite.
Public repo. No hosted demo. Last verified 2026-08-07.
Related TechTide engagement: Systems Decision Audit at TechTide AI.
Get in touch.
For speaking, collaborations, or a question about a build, send me a message. For client work, start with TechTide AI.
Send a message
I usually reply within 24 hours.