LogWeave

• 

  Analysis Middle - Event Timeline

• 

  Analysis End - Service Hierarchy View

• 

  Landing Page

• 

  Analysis Header - Root cause, Red herring

Inspiration

Engineers waste hours triaging distributed systems failures across interleaved logs from multiple microservices. And sometimes what you need is some visualization to better understand the flow and connect the dots. Why not let Gemini do its own reasoning to identify issue in seconds. Because debugging distributed systems shouldn't take hours.

What it does

LogWeave uses Gemini 3 to:

• Reconstruct chronological timelines across services
• Detect ambiguous event ordering and clock skew
• Identify root causes vs misleading symptoms
• Visualize request flows through distributed traces

How we built it

Tech Stack Next.js 14, TypeScript, Tailwind CSS Gemini 3 Flash API Prompt engineering, log parsing & trace correlation

Architecture Flow: Engineer → Upload Logs → Parser → Gemini 3 → Root Cause + Timeline + Flow → Analysis UI

Challenges we ran into

If logs are poorly construcuted it is hard to get meaningful result - so good logging should include timestamps, log level, debug logs, trace id, and span id is bonus for hierarchy view and have good context on error message.

Accomplishments that we're proud of

• Gemini identifying Red herring -- cause this is what makes most issue triaging complicated.
• Visual representation - event timeline, hierarchy -- As a software engineer I wish this service would've existed couple of years back!

What we learned

• Writing great prompt,
• Deploying demo projects fast
• Making ideas a live!

What's next for LogWeave

• One click import from company centralized log service (Elsatic Search instead manually typing)
• More integrations with other real services even before the issues happen.

Built With

• next.js
• tailwind
• typescript