Skip to content
View singularityhacker's full-sized avatar
💭
Sowing in hope
💭
Sowing in hope

Block or report singularityhacker

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
singularityhacker/README.md

Justice Conder

Software Engineer · Applied AI · Agent Infrastructure

I build software that turns complex workflows into working systems—from discovery and architecture through implementation and deployment. My background spans 15+ years across web development, enterprise delivery, and distributed protocols. Today, I focus on applied AI, agent infrastructure, and full-stack products.

Most of my production work lives in private repositories. Below are public products, technical documentation, and selected open-source components. I’m happy to discuss the architecture, implementation decisions, and engineering tradeoffs behind these systems.

Selected Work

Founded and built financial and legal infrastructure for AI agents: company formation, bank accounts, corporate records, contracts, and escrow-backed arbitration. Unified third-party services behind REST APIs, Model Context Protocol (MCP) tools, a CLI, and SMS.

Stack: Elixir, Phoenix/LiveView, Python/FastAPI, background jobs, and third-party API integrations.

September 22, 2026 snapshot: 1,633 agents · 95 USD bank accounts · 16 U.S. companies incorporated.

Product · Documentation · Adoption Metrics

An evaluation environment for AI agents operating simulated businesses. Tests long-horizon decision-making and the effects of memory, recovery, and policy constraints. Includes an open-source Python reference harness with persistent state, validated tool calls, retries, and execution boundaries.

Benchmark · Reference Harness Source

Open-source Python integration that loads ClawBank’s live MCP tool catalog into Hermes. Demonstrates dynamic tool discovery, scoped credentials, and fail-closed execution controls without hard-coding individual tool definitions.

Source Code

Agent-to-agent messaging and email, built with DarkSol. Persistent inboxes, message history, webhooks, and programmatic communication give agents a way to stay reachable across sessions.

An agent-facing marketplace for buying and selling access to private GitHub repositories. Connects skill metadata, repository access, and x402 payments into a programmatic purchasing workflow.

Elsewhere

Portfolio · LinkedIn · Writing · Email · X

Pinned Loading

  1. tiny-ai-models tiny-ai-models Public

    https://x.com/singularityhack/status/1977523296666325380?s=20

    Python 1 1

  2. webometrics-rankings webometrics-rankings Public

    Python 1

  3. cybernetic-org-design cybernetic-org-design Public

    Book Website

    CSS 1

  4. skill-shop skill-shop Public template

    Python 1 1

  5. ClawBank-co/clawbank-hermes-plugin ClawBank-co/clawbank-hermes-plugin Public

    ClawBank Hermes Plugin

    Python 1 1

  6. ClawBank-co/reference-company-harness ClawBank-co/reference-company-harness Public

    The baseline company harness for ClawBank Business Bench. Same model, same economy, 500 simulated days — the operating system is the variable. Fork it, change one policy, beat it.

    Python 1 1