# AI Startup Coach Intelligence Engine

A Python engine for analyzing early-stage startup data and generating structured coaching briefings.

This public-safe demo project shows how AI-assisted startup coaching can be made more systematic: diagnosis, risk visibility, founder prompts, and recommended next actions.

## What It Does

- Loads structured startup profile data from JSON
- Scores venture readiness across problem, market, product, GTM, revenue, team, and runway
- Detects common startup risks and execution bottlenecks
- Generates coaching priorities and founder reflection questions
- Exports Markdown and JSON analysis outputs
- Runs offline with the Python standard library

## Why This Exists

Startup coaching often becomes too generic: motivational advice, vague feedback, or disconnected frameworks. This project demonstrates a more structured approach where coaching starts from evidence and turns into focused next steps.

The engine is not a replacement for an experienced coach or advisor. It is a diagnostic layer that helps make coaching conversations sharper, more consistent, and easier to prepare.

## Quick Start

```bash
python3 -m venv .venv
source .venv/bin/activate
python -m pip install -e ".[dev]"
startup-coach examples/demo_startup.json --brief examples/briefing.md --json examples/analysis.json
```

Reference outputs are included at:

- `examples/sample_briefing.md`
- `examples/sample_analysis.json`

## Project Structure

```text
src/ai_startup_coach_engine/
  __main__.py       CLI entry point
  models.py         Typed startup profile models
  scoring.py        Readiness and risk scoring
  coaching.py       Coaching prompts and priorities
  exporters.py      Markdown and JSON exports
examples/
  demo_startup.json
tests/
  test_scoring.py
  test_coaching.py
```

## Use Cases

- Startup coaching preparation
- Founder self-assessment
- Accelerator intake review
- Venture architecture diagnostics
- AI-assisted advisory workflows
- Investor-readiness screening

## Development

```bash
source .venv/bin/activate
python -m pip install -e ".[dev]"
python -m pytest -q
```

## Roadmap

- Multi-startup batch comparison
- PDF export
- Prompt pack generation for LLM-assisted coaching
- Notion or Google Docs export
- Founder interview mode
- Stage-specific benchmark profiles

## License

MIT