Projects
Active experiments from the incubator. Each project tests a thesis about AI-powered tools that solve real process problems.
In Progress
Scope, staff, deliver, and learn from every project. A system that gets smarter with each project you close.
Thesis
Services firms have years of project data but no way to learn from it. LLMs can now extract patterns from unstructured history and use them to improve future estimates.
Signals
- •Phase 1 complete - Blueprint (scoping), Bench (talent), Retro (learnings) all working
- •Core loop functional: Scope → Staff → Deliver → Learn
- •Built in ~3 hours across 2 sessions - validates rapid development thesis
Evaluating (14)
Smart CMS
AI-powered website migration and generation
Website migration is painful and expensive. AI can understand semantic structure (not just scrape HTML) and rebuild sites that clients can actually edit.
Changelog
Auto-generated product changelogs
Product changelogs are tedious to write. Fully automatable from git - freemium for public repos, paid for private.
Brief
Vague briefs → structured requirements
Creative briefs are vague and cause rework. Conversational AI can extract structured requirements through guided questions.
Recap
Structured meeting notes with decisions & owners
Meeting notes are inconsistent or missing. Differentiate on structure: decisions, owners, deadlines extracted automatically.
Intake
Client intake calls → structured requirements
Client intake calls are unstructured and notes get lost. Transcription APIs are mature enough to extract structured requirements automatically, saving 2+ hours per client.
Drift
Contract vs reality monitoring
Contracts say one thing, reality drifts another. Document parsing + email integration can surface mismatches automatically.
Pulse
Auto-generated project status updates
'How's the project going?' requires chasing people. AI can synthesize status from existing communication and tools.
Handoff
Instant project context for new team members
Onboarding new team members to projects is painful. Document synthesis can generate context from scattered materials instantly.
Terms
Contract review that catches risks
Reading contracts is tedious and risks get missed. Document comparison can surface unusual terms against your templates.
Forecast
AI-powered revenue forecasting
Revenue forecasting is spreadsheet hell. The data already exists in tools - AI can predict cash flow automatically.
Audit
SOW vs deliverables verification
'Did we deliver everything in the SOW?' requires manual checking. Cross-document matching can verify automatically.
Docs
Keep internal docs in sync with code
Internal docs are outdated or missing. AI can watch for changes and flag when docs need updates.
Onboard
Personalized employee onboarding plans
Employee onboarding is inconsistent. AI can personalize plans based on role while ensuring nothing is missed.
Standup
Async standups without the noise
Async standups are noisy or ignored. AI can synthesize updates and surface only what matters.