Skip to content
View sergiobuilds's full-sized avatar
😀
😀

Organizations

@Ouro-labs

Block or report sergiobuilds

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.

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
sergiobuilds/README.md

Sergio builds public-grant systems and agent infrastructure

Maintainer at Ouro-labs Q00 Ouroboros stars Building GrantBack

I turn rules into working systems

I am Sergio, a Washington State CPA, three-time AI hackathon award winner, and product builder specializing in public grants and government-funded R&D. I build at the boundary between accounting judgment and software execution:

  • public-grant operations, post-award closeout, evidence review, and recovery workflows;
  • agent infrastructure, source-bound execution, orchestration, and developer tools;
  • open source, with upstream work in Ouroboros, K-Skill, ContractPlane, and Paperthin.

My recurring question is simple:

How do we turn professional rules into systems that remain inspectable, testable, and safe to act on?

Recognition

  • Open Builders Alliance Weekendthon, runner-up with SealCPA and recipient of an OpenAI-backed USD 10,000 prize
  • Build Together Hackathon, winner
  • Microsoft Copilot Hackathon, award recipient
  • Ouroboros ecosystem, open-source maintainer and upstream contributor

GitHub activity snapshot

Open-source work

Upstream contributor to the specification-first Agent OS.

  • #1387 Codex stream timeout overrides
  • #1640 per-phase interview timings
  • #1661 proposal-first onboarding contract

Maintainer across the Ouroboros product ecosystem.

Contributed a Korean business due-diligence skill suite.

  • #317 five atomic skills, one combined skill, and proxy routes
  • 2,802 additions across 30 files

Authored five upstream proposals for validation, link integrity, multi-perspective review, and intent-first agent behavior.

  • #12 SKILL.md structure and frontmatter validation
  • #13 skill-reference integrity checks
  • #14 relative-link verification
  • #15 independent perspective panel
  • #16 proposal-first intent inference

Also contributed the OpenAI-family evaluation arm to Q00/contractplane.

Products and public systems

A source-bound obligation compiler that turns fragmented post-award evidence into a candidate receivable package for accountant review.

GPT-5.6 Codex TypeScript MCP Apps SDK

Collects Korean public-funding opportunities from multiple sources and exposes them through search and MCP tools.

TypeScript Public data MCP

Cross-checks official Korean datasets for business status, pension records, sanctions, and corporate facts without adding unsupported interpretation.

Python Government APIs Evidence

A project-coherence system for coding agents built around MasterMap, Chronicle, Four Drawers, and Gatekeeper.

Shell Agent workflows Project memory

A visible AI operator that sits beside terminal AI workers.

Python Agent UX

An intent-first skill for finding the user need that a specification failed to write down.

Claude Codex Gemini Grok

Working stack

Python TypeScript JavaScript Shell OpenAI Model Context Protocol Accounting systems Public data

Contribution trail

GitHub contribution trail

Ouroboros · GrantBack · Repositories

Pinned Loading

  1. Q00/ouroboros Q00/ouroboros Public

    Agent OS: Stop prompting. Start specifying.

    Python 5.2k 520

  2. NomaDamas/k-skill NomaDamas/k-skill Public

    한국인을 위한 스킬 모음집 - SRT, KTX, 카카오톡, 한글과컴퓨터, 날씨, 미세먼지, 법령, 주식정보, 조선왕조실록, KBO, K-리그, LCK, 특허 검색, 토스 증권, 맞춤법 검사, 중고차 가격, 쿠팡, 네이버 블로그, 다이소, 올리브영, 택배 송장 조회 등등...

    JavaScript 6.5k 746

  3. LilMGenius/paperthin LilMGenius/paperthin Public

    Low-level agentic design patterns. Turning old engineering wisdom into reflexes your agent reaches for on its own—on any agent.

    Shell 115 14