Open file / 003

Public repository

Updated as the work changes

Ergin Satir / Repository

Everything public, in one place.

Not a portfolio tour.

A working index of things I build, ideas I publish, and methods I share.

Drawer / 01

Projects

Independent tools and experiments I have made public.

  1. 01Interactive decision tool

    Decision Compiler

    A transparent way to turn an ambiguous AI idea into a product role, a first test, measures, and safeguards.

  2. 02Applied AI / understanding long-form content

    Tube-Teller

    Can AI help someone decide whether a long video deserves their time—not merely create another generic summary?

  3. 03Generative AI / practical travel planning

    AI Travel Planner

    Can AI turn vague travel intent and real constraints into a plan someone can actually follow?

  4. 04AI-assisted building / public data story

    WC26 Team Values

    How quickly can AI-assisted development turn one narrow sports question into a clear public tool?

Drawer / 02

Posts

Essays written to make unfinished thinking inspectable.

  1. AI beyond ChatGPT: five roles it can play in real work

    A plain-English guide to how AI can answer, create, recommend, help decide, and act—and what each role asks from the people using it.

    Read · 6 min
  2. The model is not your moat. The loop might be.

    Five places durable AI product advantage can accumulate after model access becomes ordinary: workflow, feedback, evaluation, trust, and distribution.

    Read · 9 min
  3. AI roadmaps should start with a decision, not a model

    A practical way to move from model capability to user behavior, product loops, and durable value.

    Read · 6 min
  4. A prototype is a question, not a tiny product

    How to use AI prototypes to collapse uncertainty instead of accumulating premature software.

    Read · 5 min
  5. The AI PM's second backlog: uncertainty

    AI teams need a visible system for managing what they do not yet know—not only what they plan to ship.

    Read · 6 min

Drawer / 03

Playbooks

Practical methods designed to be used, adapted, and challenged.

  1. The one-page AI opportunity brief

    Turn a vague AI idea into a testable product bet with a named decision, observable value, and an explicit stop rule.

    Use playbook · 8 min
  2. The AI evaluation plan product teams can actually run

    Replace a vague quality score with a product-specific system for deciding whether to launch, expand, redesign, or stop.

    Use playbook · 10 min
  3. The earned-autonomy ladder for AI products

    Decide what an AI system may observe, recommend, draft, or execute—and what evidence it must earn before receiving more authority.

    Use playbook · 9 min