Personal Projects

StrataLens AI

Link: stratalens.ai

Built equity research platform using FastAPI, React, TypeScript, PostgreSQL, Redis, and Logfire for observability. Developed latency-optimized agentic RAG system using Qwen3-235B that synthesizes earnings calls, SEC filings, and real-time news via Tavily to answer complex financial queries. Built SEC filings RAG pipeline achieving ~85% accuracy on FinanceBench benchmark with LLM-as-judge evaluation. Developed text-to-SQL stock screener and market data analysis tool using DuckDB for natural language financial queries.

Financial Podcast Platform

Built real-time AI podcast platform using OpenAI Whisper and GPT-4o with Celery workers for async audio generation. Developed React frontend with WebSocket streaming for live audio updates, deployed on AWS. Organically acquired 50+ users in early beta with portfolio-based personalized podcast generation on-demand. Implemented market data pipeline processing 9,000 tickers every 20 minutes using SERP API and Redis caching.

TheMarketCast.ai

Link: themarketcast.ai

Built platform parsing and analyzing SEC Form D filings daily, providing accessible insights into private capital formation trends across U.S. markets using Python and SEC EDGAR API.

Jobs Intelligence Scraper

Built large-scale web scraper using Playwright and Meta-Llama-3-70B to extract 100,000+ job postings from high-growth startups and S&P 500 companies with structured data extraction and analysis capabilities.

GreyNSights

Link: github.com/kamathhrishi/GreyNSights

Framework for privacy-preserving data analysis using Pandas with pointer-based architecture for flexible EDA. Implemented differential privacy for individual row protection and federated analytics using secure multi-party computation.

Open Source Contributions

PySyft, OpenMined

Links: github.com/OpenMined/PySyft • github.com/OpenMined/SyMPC

Core Contributor (2019-2022). Performed code reviews and planned SyMPC library roadmap for privacy-preserving ML framework. Implemented FALCON protocol operations - first Python implementation of honest-majority maliciously secure framework for private deep learning.

Key Contributions:

  • FALCON Protocol: First Python implementation of honest-majority maliciously secure framework
  • SyMPC Library: Planned roadmap and performed code reviews for secure multi-party computation library
  • Framework Development: Contributed to core privacy-preserving ML infrastructure using PyTorch, TensorFlow, and differential privacy