Hi, I'm
Sidessh More
About Me
Hi there! 👋 I'm a developer who enjoys building meaningful solutions. Whether it's designing intuitive apps, automating workflows, or experimenting with new technologies, I love taking on challenges that push me to learn and grow. For me, technology is a tool to solve real problems, and I'm always excited to explore what's next.
GitHub Contributions
Technical Skills
Experience
- Engineered FastAPI-based backend services with RAG-enhanced LLM routing, implementing AWS-deployed ML pipelines and CI/CD workflows for research query processing
- Implemented chat-history retrieval system integrated with Supabase, injecting conversational context into LLM calls to improve response relevance across research experimentation
- Architected cross-platform mobile application using Flutter with Supabase backend, integrating secure auth and real-time data sync to serve 500+ rural women with health education
- Designed accessibility-first UI with multi-language support and voice navigation across 15 app screens, collaborating with designers through iterative testing
- Developed RESTful APIs using Django, employing caching and query optimization to achieve 60ms response time for ERP dashboard operations
- Led initiative to restructure PostgreSQL schema across 15+ enterprise tables, deploying indexing strategies to reduce latency for high-volume data operations
Featured Projects
Referrlyy
Full-stack referral networking platform with cross-platform mobile and web architecture, enabling verified professional connections at scale.
Technical Highlight
Architected a scalable referral-matching system using PostgreSQL with indexed queries, ExpressJS middleware for request validation, and Flutter's BLoC pattern for state management. Deployed on Render with CI/CD pipelines for automated testing and deployment.
BoltPrep
Real-time AI interview platform with speech-to-text pipeline, LLM-powered evaluation, and performance analytics engine.
Technical Highlight
Engineered low-latency audio processing pipeline combining Deepgram's streaming STT with Gemini API for contextual answer evaluation. Implemented efficient state management using Riverpod with caching strategies to minimize API calls while maintaining real-time responsiveness.
Aqua Trace
ML-powered water footprint tracking app with custom CNN model for food classification and consumption analytics.
Technical Highlight
Implemented transfer learning on MobileNet V2 with custom classification head, applying data augmentation and fine-tuning to achieve 92% accuracy. Converted to TFLite with INT8 quantization for efficient mobile inference under 100ms.
Mili
Mental wellness platform with LLM-powered conversational AI, mood analytics, and crisis detection system built at SunHacks.
Technical Highlight
Designed efficient context management by chunking conversations into summaries before LLM calls, reducing token consumption by 60% while preserving conversational continuity. Implemented PostgreSQL-backed mood tracking with time-series queries for trend analysis.
Auto EDA
Automated exploratory data analysis pipeline supporting 100K+ row datasets with intelligent visualization selection.
Technical Highlight
Built modular analysis pipeline using Pandas for data profiling, automatically detecting numeric/categorical columns and applying appropriate statistical tests. Implemented lazy loading with Streamlit caching for handling large datasets efficiently.
Blog
View all postsLoading posts...