Work

Three projects. & a clear pecking order.

JustAi is the public project and demo. Agent-Inc is the one that taught me how to get there. The harness is the one I use every day, with the *-mini repos pulled out as public references.

Public proof path · live demo · JustAi source · substrate sketch

Flagship · Public demo

JustAi

An enterprise observability layer for multi-agent engineering teams.

Six-stage pipeline — intent, plan, execute, review, synthesize, report — wrapped in cost, latency, and learning dashboards. Trajectory-based evaluation means every run becomes training signal for routing decisions. Built on SpacetimeDB, LiteLLM, LangFuse, and a thin React reporting layer.

source
Public repo and live demo available
dashboards
6 mission control through agents
operator loop
Trace-first cost, latency, review, memory

Enter the case study

2026 · Astro · TypeScript · SpacetimeDB · LiteLLM · LangFuse · Recharts

JustAi Observability — cost, latency, and quality from LangFuse traces
Observability · cost, latency, and quality from LangFuse traces

Prior work · 2024

Agent-Inc

A tri-agent harness built on intuition, before the literature caught up.

Observer / Analyzer / Actor, coordinated by an orchestrator that handled intent, schedule, handoff, and trace. It ran. It also taught me that handoff and memory are the load-bearing layer — more than whichever prompt was trending that month. That lesson is now encoded into JustAi as first-class infrastructure.

What I built, what broke, what I learned

2024 · Python · OpenAI · local tools

Agent-Inc tri-agent architecture Three agents — Observer, Analyzer, and Actor — connected in sequence, with an orchestrator coordinating them and a feedback loop from Actor back to Observer. Agent 01 Agent 02 Agent 03 Observer Analyzer Actor Watches · listens Plans · reasons Writes · executes context plan Orchestrator · coordination plane Intent › schedule › handoff › trace Feedback · outcome signal

Public substrate

Public substrate

Minimal public references for the moving parts of an agent harness.

The *-mini family — minimal, MIT, well-tested public references for the moving parts of an agent harness. safe-mini (safety reference), lab-mini (data-science labbing), route-mini (multi-provider LLM routing with fallback), memory-mini (durable agent memory).

Read the harness sketch

2026 · MIT public references · active substrate