Intelligent SQL query router with cost-based optimization, automatic backend selection (DuckDB/Polars/Spark), and partition pruning for 50-100x speedups

Optimized PostgreSQL queries on a computer science publications dataset. Created tables, ingested CSVs, and wrote queries to analyze conferences, authors, and publication trends. Improved performance by designing indexes, refining join/filter logic, and evaluating execution plans with EXPLAIN, demonstrating query tuning and indexing strategies.

An advanced distributed knowledge fabric for intelligent document processing, featuring multi-document agents, optimized query handling, and semantic understanding.

[BSc Thesis] PriSM: Prior-Guided Search Methods for Query Efficient Black-Box Adversarial Attacks

Django Learning Journey: A structured guide to learning Django step by step. This resource includes detailed notes, examples, and practical exercises covering models, ORM, database interactions, and core Django concepts. It is designed for hands-on learning and to serve as a personal reference

LIMAO is a Lifelong Modular Reinforcement Learning framework designed for database query optimization in dynamic environments.

🪂 Parachute: Single-Pass Bi-Directional Information Passing (VLDB'25)

A Dockerized Online Shop API built with Django REST Framework

[TIP25] Code for "Text-Video Retrieval with Global-Local Semantic Consistent Learning"

Analyzed Spotify’s 12M-song dataset using SQL and NoSQL modeling, normalization, indexing, and the Apriori algorithm for pattern mining.

Instructional materials (course files) for the BBT3104 course (Advanced Database Systems) and the MIT8107 course (Advanced Database Systems). Topic: Query Optimization. Based on the IMDb dataset using PostgreSQL and the Join Order Benchmark (JOB).

Case Study for a data engineering job application at a company

Sub-optimal Join Order Indicator

torch-like convolution over binary trees

This repository presents research on SQL query optimization, exploring how different indexing strategies and PostgreSQL configurations impact query performance. The project, which earned second place at TCCD Research Day, provides insights into improving database efficiency.

PostBOUND is a framework for studying query optimization algorithms for (relational) database systems.

platform for benchmarking hint-based optimisation approaches

AI-driven solution that allows users to filter and retrieve tickets based on natural language queries. This system interprets and translates user inputs into technical filter criteria, providing an intuitive, user-friendly interface for effective ticket management.