ProductOS
The team that thinks before it builds.
PM → Sys Arch → SW Arch → Launch Eng
Model & API Key No API key — click to add yours before building
▾
Model
API Key
Get your key at console.anthropic.com · stored only in your browser, never sent to our servers
How to set up Ollama — free, runs on your machine
01
Install Ollama Download and install from ollama.com — available for Mac, Windows, Linux
02
Start Ollama Open the Ollama app — or run in your terminal: ollama serve
03
Pull a model (once) In a new terminal tab, download the model you want: ollama pull llama3.2 Other options: mistral · gemma2:9b · qwen2.5:14b
04
Select model above & hit Build No API key needed — ProductOS connects to localhost:11434
Only works when ProductOS is running locally on the same machine as Ollama. Quality varies by model — Llama 3.2 is a good starting point.
Key stored in your browser only
3 quick questions · < 60 s
Who is this for? Why: shapes UX, pricing, and how users discover your product. If you don't know yet, pick "Not sure" — the PM agent will figure it out.
Free or paid? Why: sets the business model your PM starts from. No wrong answer — freemium is a valid hedge if you're torn.
Launch timeline? Why: tells the architect how much to simplify the initial design. Fast means cut scope aggressively. No rush means build it right.
● Error
01 PM PRD
02 Sys Arch HLD
03 SW Arch LLD
04 Launch Eng Plan
IDLE
01 Product Manager
PRD
WAITING
▾
Submit an idea to begin
02 System Architect
HLD
WAITING
▾
Waiting for PRD
03 SW Architect
LLD
WAITING
▾
Waiting for HLD
04 Launch Engineer
Plan
WAITING
▾
Waiting for LLD
All 4 artifacts ready
Push to GitHub
Repository URL
Personal Access Token Needs public_repo scope for public repos, repo for private. Generate one at github.com → Settings → Developer settings → Personal access tokens.
Pushing…
Session History
No sessions yet