Tech Stack
Next.js
React
Tailwind CSS
Node.js
Typescript
tRPC
TanStack Query
Drizzle ORM
PostgreSQL
Google GenAI SDK
Zod
Description
This project bridges the gap between static career advice and personalized guidance. Using the Google GenAI SDK, I built a counselor that maintains context over long conversations to provide tailored advice.
A key technical challenge was ensuring session continuity. I designed a scalable PostgreSQL schema using Drizzle ORM that achieves 99% accurate session retrieval, allowing users to pause and resume conversations seamlessly.
- Developed a full-stack AI counseling platform handling 100+ chat interactions per session.
- Engineered a human-like chat interface with live typing indicators and streamed responses, reducing perceived latency by 40%.
- Designed a scalable PostgreSQL schema using Drizzle ORM for robust data persistence.
- Implemented context-aware responses using Google Gemini and tRPC for type-safe API calls.
Page Info
Chat Interface
Human-like chat interface with typing indicators and streamed AI responses.
Instant Session Reset
Effortlessly clear complete conversation history with a single action, ensuring user privacy and enabling fresh context starts.
Adaptive Dark Mode
High-contrast dark theme optimized for visual comfort, reducing eye strain during extended counseling sessions.
