Daily Pilot

Inspiration

Most productivity tools focus on capturing tasks, but very few help with the harder problem: deciding what actually matters today. The inspiration behind Daily Pilot came from personal frustration with juggling many responsibilities while constantly context-switching. I wanted a tool that behaves less like a checklist and more like a calm thinking partner—one that helps reason through priorities, trade-offs, and intent before jumping into execution.

What it does

Daily Pilot is an AI-assisted daily planning companion.

Instead of asking users to pre-structure everything into rigid fields, it encourages natural-language planning. Users can describe their day, constraints, or goals, and Daily Pilot helps:

• Surface priorities
• Identify conflicts or overload
• Suggest a more intentional plan for the day

The emphasis is on decision clarity, not task automation.

How we built it

The app is built as a lightweight web experience using:

• React for the UI
• TypeScript for type safety and maintainability
• Vite for fast iteration
• Gemini 3 as the intended intelligence layer
• Vertex AI / Google Cloud as the target production setup

The internal design focuses on short conversational loops, explainable suggestions, and user-controlled iteration rather than opaque “auto-planning.”

Challenges we ran into

The biggest challenge was model accessibility.

Although Gemini 3 is central to Daily Pilot’s vision, the models are currently in preview and not reliably usable via Vertex AI, even on paid plans. The Public Gemini API does work, but the available quota is too limited for sustained development, testing, and iteration—especially for conversational workflows.

This constrained deeper experimentation with agent-style reasoning and forced the project to remain more architectural and conceptual than originally planned.

Accomplishments that we're proud of

• Designing a planning experience that feels thoughtful rather than prescriptive
• Building a clean, extensible frontend architecture
• Clearly separating UI, reasoning flow, and future AI layers
• Staying transparent about limitations while still delivering a coherent product direction

What we learned

• Productivity is as much about decision support as task tracking
• AI assistance is most valuable when it explains why, not just what
• Access to production-grade AI infrastructure matters a lot for developer velocity
• Even limited AI access can help validate user experience and interaction patterns early

What's next for Daily Pilot

Once stable, full-quota access to Gemini 3 (or similar models) is available, the next steps include:

• Richer reasoning chains and multi-step planning
• Memory and context across days
• More explicit trade-off explanations
• Experimenting with proactive vs user-initiated guidance

Daily Pilot is best viewed as an exploration of AI-assisted thinking for daily life, with a clear path toward becoming a practical, everyday companion.

Built With

• gemini3
• google-cloud
• react
• typescript
• vertexai
• vite