Current Projects

Multi-Agent Data Analytics

Agentic AI workflows to automate data science tasks

Prototyping multi-agent systems that coordinate intelligent workflows to clean data, perform exploratory analysis, uncover causal relationships, and build predictive models across complex datasets.

Tech: LangChain, LangGraph, Python (async workflows, agents, orchestration), scikit-learn, PyTorch, causal inference libraries, statistical modeling frameworks, data engineering tools (Pandas, Polars).

Financial Advisor

Personalized financial guidance and practical tools

Features: Mortgage calculator, retirement planner, network calculator, goal tracking, document analysis, payment reminders, conversation management

Stack: React, Flask, OpenAI, Python

Domain-Specific AI Models

Fine-tuned models for contextual understanding

Training and fine-tuning specialized AI models to enhance contextual understanding and deliver more accurate, domain-aware recommendations.

Focus: Model fine-tuning, Domain adaptation, Contextual AI

Built with OpenAI Agents SDK

Career Agent

Answers user questions about me and sends push notifications (via Pushover) for unknown questions and user outreach.

SDR Agent

Generates multiple sales emails in various tones, selects the most effective one, formats and sends to specified email addresses.

Deep Research Agent

Takes user queries, plans multiple related searches, executes them concurrently, writes comprehensive reports, and emails the results.

Built with CrewAI

Debate System

Two agents argue for and against a motion with a judge agent to decide the winner.

Financial Researcher

Research crew for comprehensive topic analysis and reporting using Serper for web search.

Stock Picker

Identifies trending companies, researches them, and selects the best investment opportunities.

Coder

Uses CrewAI Coder Agents and Docker to write and execute code autonomously.

Engineering Team

Multi-agent team including engineering lead, backend engineer, frontend engineer, and test engineer that collaboratively designs, codes, and tests according to user requirements.

Trading Simulation Platform

Built with the Engineering Team by providing specific requirements for a complete trading simulation system.

Built with LangGraph

Autonomous AI Assistant

Developed an autonomous AI assistant system using LangGraph and LangChain that iteratively completes user-defined tasks through a worker-evaluator architecture.

Integrated Tools: Playwright web automation, web search APIs, Python execution, file management, with a Gradio interface for user interaction.

Interested in collaboration? Get in touch →

Ask me anything