Skip to content
View benmfzen's full-sized avatar

Block or report benmfzen

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
benmfzen/README.md

Benjamin Zengler

Berlin · ex-COO of Fyrfeed (acquired 2024) · founder at ai1 Ventures

Right now: a fully autonomous content pipeline shipping short-form video daily (research → script → render → schedule, built on Claude Code), an installable MCP server for running AI transformation programs, and the repos below — each one built to demonstrate the same thing a different way.

The pattern across all of them: agents that must cite their sources, pass tests in CI, and refuse rather than guess when they can't back a claim. That's not a slogan restated per repo — it's what the numbers below are.

The repos, and why they exist
postpeer-pilot A performance-driven publishing autopilot for short-form video, as an MCP server. 16 reliability/invariant tests, a measured 18/18 agent-eval suite (incl. 9/9 unsafe-action refusals — the agent tried to schedule live, the tool layer refused), and a backtest that reports honestly where its own conservative design costs performance, not just where it wins.
meta-youtube-comment-mcp Human-in-the-loop comment automation for Instagram, Facebook & YouTube. 47 deterministic tests, a written threat model, an eval harness that measures classifier + routing accuracy on labeled data — not just vibes.
ai-transformation-90 A transformation copilot for the first 90 days of an AI program — playbook, costed business cases, working prototypes, and the method itself as an MCP server with evidence-gated scoring. Swap in your own org and it runs your program.
know-no-hearsay A reference architecture for evidence-bounded generative media: no pipeline step may increase the epistemic strength of a claim without evidence. snakeoil-radar (finds viral health-misinformation claims on TikTok, checks them against PubMed) and cited-cuts (turns a found claim into a quote-bounded rebuttal video) are two applications of it.
sorting-hat An LLM-classified document inbox: drop scans into one folder, get named, filed paperwork out. Boring, useful, runs daily on my machine.
inanimatus LLM-driven CAD, done declaratively — geometry-tested CadQuery models from dimensioned specs, as a documented method + Claude Code skill.

📫 LinkedIn · 🎬 the daily-shipping pipeline in action: @dabigredbutton

Pinned Loading

  1. ai-transformation-90 ai-transformation-90 Public

    A transformation copilot for the first 90 days of an AI program — playbook, opportunity portfolio, costed business cases, operating model, plus the software to run it: MCP assistant with evidence-g…

    Python

  2. inanimatus inanimatus Public

    Conjuring solid objects from precise incantations — a declarative method for LLM-driven CAD (CadQuery + Claude Code skill)

    Python

  3. know-no-hearsay know-no-hearsay Public

    A reference architecture for evidence-bounded generative media pipelines — grounded, eval-gated, with a fail-closed-to-publish model. Curated extract; ships no third-party data.

    Python

  4. snakeoil-radar snakeoil-radar Public

    🐍 A reaction radar for viral health misinformation — a Claude Code skill. Finds high-reach false claims on TikTok and checks them fail-closed against real PubMed studies. Free, no API keys.

    Python

  5. sorting-hat sorting-hat Public

    An LLM-classified document inbox: drop scans into one folder, get named, filed paperwork in plain folders. No server, no database, no OCR stack.

    Python

  6. postpeer-pilot postpeer-pilot Public

    Performance-driven publishing autopilot for short-form video, as an MCP server on top of the Postpeer API

    Python