Skip to content
View DiscoStew6082's full-sized avatar

Block or report DiscoStew6082

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
DiscoStew6082/README.md

Stewart Emberton

AI/LLM systems builder focused on grounded generation, eval gates, provenance, local-first agents, and accessibility-aware workflows. Former Microsoft Escalation Engineer.

Public proof

Groundball — eval-gated local AI over structured data

Groundball is a local-first baseball analytics assistant that answers MLB history questions with DuckDB-backed facts, visible SQL, provenance metadata, stat-claim verification, and deterministic release gates.

What it demonstrates:

  • Grounded generation: LLM prose is constrained by typed routing, source rows, and dataset manifests.
  • Eval-gated releases: deterministic CI checks guard supported questions, unsupported cases, SQL visibility, and provenance coverage.
  • Audit-ready answers: responses expose SQL templates, row evidence, checksums, license metadata, route metadata, and review queues.
  • Local-first AI workflow: Python, FastAPI, DuckDB, Typer, Gradio, pytest, and local/open-weight model integration.

Repo: https://github.com/DiscoStew6082/groundball

YouTube Channel Searcher and Playlist Maker — local-first metadata search

A small browser app that imports one channel's public YouTube metadata through the official API, stores it locally in SQLite, searches titles/descriptions/tags, and builds temporary playlist queues from matching results.

What it demonstrates:

  • Privacy-conscious integration: no scraping, video downloads, transcripts, or account access.
  • Useful product loop: import, search, inspect, queue.
  • Tested local app: FastAPI, Jinja, SQLite, httpx, pytest, and GitHub Actions.

Repo: https://github.com/DiscoStew6082/youtube-channel-searcher-and-playlist-maker

ObviousSaveMD — native macOS Markdown editor

A SwiftUI/AppKit desktop editor that keeps Markdown as plain text, builds a Finder-launchable .app, and treats hostile file intake as a first-class design boundary.

What it demonstrates:

  • Native macOS UX: SwiftUI app shell with AppKit NSTextView editing behavior.
  • File-safety boundaries: rejects symlinks, special files, oversized files, NUL bytes, and invalid UTF-8 before loading content.
  • Tested desktop edge cases: Swift Testing coverage for document state, stale async opens, save flows, and paste behavior.

Repo: https://github.com/DiscoStew6082/ObviousSaveMD

Current focus

Building practical AI systems that make model behavior inspectable: grounded answers, reliable fallbacks, provenance, evals, accessibility, and local-first automation.

Pinned Loading

  1. groundball groundball Public

    Eval-gated local-first baseball analytics assistant using DuckDB, provenance, stat-claim verification, and deterministic AI release gates.

    Python

  2. youtube-channel-searcher-and-playlist-maker youtube-channel-searcher-and-playlist-maker Public

    Local-first YouTube channel metadata search and playlist queue builder using FastAPI, SQLite, and the official YouTube Data API.

    Python

  3. ObviousSaveMD ObviousSaveMD Public

    Native macOS Markdown editor built with SwiftUI/AppKit, conservative file intake checks, and plain-text local editing.

    Swift