I'm a Machine Learning Engineer specializing in Quantitative Finance, Industrial AI, and production ML systems. Based in New York, I build systems that combine high-performance computing with advanced ML to solve real-world problems — from predicting market drawdowns to detecting industrial faults in real time.

class AtharvaJoshi:
    def __init__(self):
        self.location = "New York, USA"
        self.current_focus = ["Quantitative Finance", "Low-Latency Systems", "AI Agents"]
        self.interests = ["Market Microstructure", "Predictive Maintenance", "Neural PDEs"]
        self.languages = ["Python", "C++", "SQL", "TypeScript"]

    def say_hi(self):
        print("Let's build something amazing together!")

me = AtharvaJoshi()
me.say_hi()

Universal multi-agent simulation for predicting outcomes

• Swarm-based optimization and prediction through emergent collective behavior
• Multi-agent simulation engine with configurable agent strategies
• Applicable across finance, logistics, and complex systems modeling

Tech: Python · Multi-Agent Systems · Swarm Intelligence · Simulation

Predicts cash crises 90 days before they hit

• ML-powered forecasting of personal financial distress signals
• Full-stack application with React frontend and FastAPI backend
• CI/CD pipeline with Docker containerization
• Free, private, and open source

Tech: Python · FastAPI · React · Docker · GitHub Actions

High-Performance Trading Infrastructure with Predictive Execution

• Sub-microsecond order book operations with lock-free data structures
• ML-powered trade flow prediction using gradient boosting
• C++20 core engine with Python ML pipeline
• Memory-pooled architecture for zero-allocation hot paths

Tech: C++20 · Python · XGBoost · Lock-free Programming · SIMD

Production-grade derivatives pricing using deep learning

• Deep Galerkin Method for solving Black-Scholes and exotic option PDEs
• Handles high-dimensional pricing problems intractable for finite differences
• Comprehensive visualization of option surfaces and Greeks

Tech: PyTorch · Neural Networks · PDEs · Quantitative Finance

99.98% accuracy vibration fault detection

• 8-class combined fault detection system with sub-second inference
• Lightweight model (1.7 MB) deployable on edge devices
• Real-time monitoring dashboard with Streamlit

Tech: MiniRocket · scikit-learn · Streamlit · Signal Processing

Personal ML/AI portfolio with interactive demos

• Showcases projects with live demonstrations and case studies
• Modern responsive design with Next.js

Tech: Next.js · React · TypeScript · Tailwind CSS

Languages

ML & Data Science

Infrastructure & Tools

Web & Visualization

I'm open to collaborating on projects in:

• Quantitative Finance & Trading Systems
• Machine Learning & Deep Learning
• High-Performance Computing
• Industrial AI & Predictive Maintenance