Erdeniz TunçProduct Engineer
WorkAboutHow I use AIWritingContact

How I use AI

AI helps with specific steps. I make the product decisions.

How each one runs

Interview recordings

Optivideo · Optifeed

  1. InputRecorded internal interviews.
  2. AINotebookLM transcribes each recording.
  3. AIClaude Code reads transcript plus prototype code.
  4. OutputA TODO.md of needed changes.
  5. I decideI keep, cut or defer each item.
  6. NextBuilt. The next feedback round tests it.

Seen in Optivideo

Behaviour data

This portfolio

  1. InputPostHog events and funnels.
  2. AIClaude Code or Codex reads them. Next to the site's code.
  3. OutputA change list, each tied to data.
  4. I decideI pick what ships.
  5. NextBuilt. Measured again in PostHog.

Product feedback

My own work · an app clone

  1. InputUser feedback and research.
  2. AIGroups the repeated problems.
  3. I decideI name the problem worth solving.
  4. AIReads the flow and code. Proposes changes.
  5. NextI build and test. Not just recommend.

An untested idea

Product development model · Optifeed

  1. I decideNine-field PRD. One Key User Action.
  2. AIA coding agent builds a frontend prototype. In a day or two.
  3. InputInterviewees react to it. CTO checks feasibility.
  4. AIWrites the Product Knowledge Document. From prototype code and notes.
  5. I decideI set V1's inputs. AI formats scope, never chooses.
  6. NextPhased, spec-driven build. Or engineering handoff.
Prompt: Product Knowledge Document
You are a product architect. Read the codebase in this directory, thoroughly source files, types, API routes, state management, data structures config files, and any existing documentation. Also read the discovery notes provided below.
Then produce a single, comprehensive Product Knowledge Document in Markdown.
The document must be written as if for a new team member or an AI agent that
will work on this product. It must reflect only what actually exists in the code
and what was confirmed during discovery. Do not invent features, do not speculate
about future plans.
Required sections:
1. Goal: one sentence — what does this product do and for whom?
2. Core Questions: 3 to 6 questions the product answers for its user
3. Product Definition: 2 to 3 paragraphs. What it is, what it is not.
4. Current Problem / Target Experience: side-by-side comparison
5. End-to-End Product Workflow: ASCII block diagram of the full flow
6. Key Decision Logic: branching and routing logic as ASCII decision trees
7. Data Model: core TypeScript interfaces with field-level explanations
8. State Management: how state flows through the product
9. Scoring or Classification Logic: criteria and weights as a table, if applicable
10. Invalid / Weak Signals: what the product explicitly ignores and why
11. Output / Export: what the product produces for the user
12. V1 Scope: In Scope / Out of Scope as two lists
13. Open Questions: unresolved decisions visible in the code
14. Known Limitations: gaps, edge cases, honest weaknesses
15. Success Criteria: how you would know the product is working
For every major decision in the product, add a line:
"Design principle: [why this choice was made]."
Discovery notes:
[PASTE NOTES HERE / Or tell the coding agent which md files to read]
Prompt: V1 scope
You have the product knowledge document below.
Your task is to define V1.
V1 must deliver the core user flow and the core value proposition.
Nothing more.
Use the following inputs to scope V1:
CTO decisions and constraints:
[PASTE CTO DECISIONS]
Core requirements from the discovery loop:
[PASTE DISCOVERY REQUIREMENTS]
Must-haves for V1:
[PASTE MUST-HAVES]
Using the knowledge document and the inputs above, produce:
1. Core User Flow: the path from entry to core value, step by step
2. V1 Scope: what is in, what is out, as two lists
3. No-Gos: what is deliberately excluded from V1
4. Success Signal: how you will know V1 is working
[PASTE KNOWLEDGE DOCUMENT / Or tell the coding agent which md files to read]

Seen in the product development model, Optivideo

A manual job

OptiGTM · Optifeed

  1. InputMarketing researched leads by hand.
  2. I decideFlow mapped in Miro. Scope agreed with the CTO.
  3. AIBuilt in Rails. With Claude Code and Codex.
  4. AIClaude explains each candidate company.
  5. OutputResults saved per project.
  6. I decideMarketing reviews before any outreach.

Seen in OptiGTM

What all five share

Different inputs

  • Interview recordings
  • Behaviour data
  • Product feedback
  • An untested idea
  • A manual job

Different AI work

  • Transcribe
  • Read code and data
  • Cluster evidence
  • Prototype
  • Write specs and tasks
  • Build

The same principles

  1. Evidence before prompts.
  2. AI works in real product context.
  3. Every AI step produces an artifact.
  4. Build and test. Don't just discuss.
  5. I make the product decisions.
  6. Real behaviour feeds the next loop.
  7. One step's output feeds the next.

Who does what

AI

  • Transcribes and structures
  • Reads code and data
  • Offers alternatives
  • Turns findings into tasks
  • Builds bounded changes

I decide

  • Which problem matters
  • What evidence is enough
  • What to build, and not
  • Whether it ships

Full idea-to-V1 workflow: How I work as a Product Engineer →

Email me →What I am open to →