Projects
Have a look at some of my personal projects. Checkout my GitHub for more.
Adaptive Database Optimizer
2025README
A self-tuning system that watches SQL workloads and automatically retunes how Parquet data is partitioned and sorted, routing traffic across layout versions with multi-armed and contextual bandits (UCB1, Thompson sampling) and keeping only statistically confident improvements.
BanditsAdaptiveRoutingAutomationPartitioningDuckDB
QueryGPT: NL-to-SQL Pipeline
2025README
A multi-agent NL-to-SQL pipeline of four staged agents (intent, table selection, column pruning, SQL generation) that progressively narrows schema context to cut token usage, with pydantic-ai structured outputs and automatic retries, llama-index + Neo4j hybrid vector store RAG, and end-to-end SQL validation.
LLMsStructured outputsRAGllama-indexpydantic-aiNeo4j