Skip to content
View BytesFromToby's full-sized avatar

Block or report BytesFromToby

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
BytesFromToby/README.md

Toby Iverson

Automation engineer in Minneapolis. I build AI systems that check their own work — agentic workflows, LLM integration, guardrail design. Python · SQL · FastAPI.

Verification beats trust. Software should be able to prove it did what it claims.

Shipped

  • RepoRouter — autonomous GitHub issue-triage operator, live in production on a local 8B model. Safety enforced by the harness; guardrails pinned by pytest.
  • Polis — strategy game where 28 AI factions negotiate through live LLM audiences; a deterministic engine pinned by 638 tests enforces every deal.
  • Plumbline — spec-driven workflow: eight single-responsibility agent skills and two unattended orchestrators take a written brief to verified code. Cross-skill contract audited in CI.
  • Log Analyzer — desktop log analysis in pure Python 3 + Tkinter, zero external dependencies.

Elsewhere

🌐 Portfolio · 💼 LinkedIn

Pinned Loading

  1. RepoRouter RepoRouter Public

    Autonomous issue-triage operator — checks every new issue against your docs, then replies, labels, closes, or escalates. Runs live in production on a local 8B model; safety enforced by the harness,…

    Python 1

  2. plumbline plumbline Public

    A "trust, but verify" workflow of AI-agent skills: prompt to verified code, every step trackable and deviations logged — plus an autonomous maintenance loop to keep a built project honest.

    Python 3

  3. Polis Polis Public

    Govern the ungovernable: 28 AI factions negotiate through live LLM audiences; a deterministic, 638-test engine enforces every deal.

    Python 2

  4. Log_breakdown Log_breakdown Public

    Desktop log analysis — merge multi-file logs in timestamp order, filter noise, match known issues, build reports. Pure Python 3 + Tkinter, zero external dependencies.

    Python 3 1