Proud to present my own personal project Yeet, a peer to peer file sharing app over the network. Yeet is a cross-platform file sharing application built with Rust (Tauri backend) and React (TypeScript frontend) that enables seamless file transfers between devices on the same network. The app uses a dual-protocol approach with QUIC for control messages and HTTP for actual file transfers. The application includes real-time progress tracking, global hotkeys for quick access, and supports adaptive chunk sizing with parallel downloads for optimal performance.
22
130
Last night, I started an interesting offer from a client. They wanted an ai chatbot that analyzes pdfs and their wiki data. So far, I have managed to create a chatbot that ingests PDFs and internal wiki data using LangChain + Vector DB (Pinecone/Weaviate). As a result we exposed a REST + UI interface for querying via natural language.
15
96
Designed a powerful and high-performance analytics dashboard for an e-commerce site that can ingest and visualize millions of real-time transactions. The back-end was developed in Rust, chosen for speed and memory safety, and responsible for data ingestion and aggregation. Data was persisted to a horizontally scalable NoSQL distributed database system (Cassandra) to ensure operational uptime. Data pipelines were designed with Kafka for rendering efficient streams of processing. The front-end is built in Solid.js and TypeScript and offer interactive charts and custom reports. Used a framer template for the website (separate from the app itself).
25
146
Working on High-Performance Real-Time API with WebSockets & Redis
Progress (1/4): Built a low-latency real-time communication layer for a live dashboard system using Go + WebSockets and Redis Pub/Sub. Achieved sub-50ms message latency under 50k concurrent users.
15
88