projects and competition highlights
Award-winning voice-driven AI chatbot for humanoid robots; engineered using tech stacks like LangChain & AWS-stacks (Transcribe, Polly, EC2, Bedrock) with infrastructure managed via Terraform.
Collaborated on backend with Gin & GORM(PostgreSQL) while navigating a large open-source codebase.
Build a system that checks whether a claim is supported or refuted by a fact database, then returns the evidence sentences.
Predict how small molecules change gene expression in different cell types. The competition is part of the NeurIPS 2023 Competition Track.
The goal of this competition is to predict e-commerce clicks, cart additions, and orders. We build a multi-objective recommender system based on previous events in a user session with Matrix Factorization with Pytorch.
The goal of this competition is to predict sales for a chain of stores. We build a time series forecasting model with LSTM and XGBoost.
A website for visualizing what I consumed that were worth it.
Automated user badge validation for gift tier claims and was used over 1,500 times by community members during the AI Study Jam period.
A basic and intuitive Python module for making sense of (Vector Space) IR system.
A command-line tool for managing and organizing prompts for AI assistants. Equipped with fuzzy search.
A LINE chatbot powered by Google Gemini API for conversational AI.
Educational resources and examples to understand MapReduce concepts. Use Docker to simulate master and worker nodes.