Engineering Undergrad ('22β'26) | AI/ML Engineer & Full-Stack Developer
I bridge the gap between complex machine learning research and high-performance production systems. My focus is on building scalable, data-driven tools using Python, Go, and FastAPI.
- High-Concurrency Systems: Engineered a Go crawler capable of processing 10k+ URLs in <5 seconds.
- Scalable AI Pipelines: Built a FastAPI/React ecosystem that parses and analyzes 50MB+ PDFs in <10 seconds.
- Anomaly Detection: Developed a GMM-based fraud model achieving a 0.93 ROC-AUC on 280k+ transactions.
Current Interests: NLP & LLM Orchestration, Computer Vision, and Medical Imaging.
| Category | Tools & Technologies |
|---|---|
| AI/ML |     |
| Systems/Backend |     |
| Cloud & DevOps |     |
| Frontend/Data |    |
Contract Intelligence Parser Full-stack Document AI for massive PDF analysis
- Performance: Processes 50MB+ documents in under 10 seconds with 95% extraction accuracy.
- Stack: FastAPI, React, MongoDB, Docker.
- Links: π₯ Live Demo β’ πΈ Screenshots
High-Performance Go Crawler Concurrent scraper optimized for ML data pipelines
- Performance: Scrapes 10k+ URLs in <5 seconds using Go's concurrency primitives (Goroutines/Channels).
- Stack: Go (Golang), Colly, ML-ready JSON output.
- Links: π οΈ View Source
Unsupervised Fraud Detection Anomaly detection on high-dimensional financial data
- Results: Achieved 0.93 ROC-AUC on 280k+ transactions using Gaussian Mixture Models (GMM).
- Stack: Python, Scikit-Learn, NumPy, Matplotlib.
- Links: π Kaggle Notebook
Iβm driven by the challenge of turning complex research into production-ready tools. I build with a global mindset and a passion for scalable innovation.
Off-duty? Iβm your unofficial γγ£γ³γ©γ³γ観ε 倧使. Letβs venture into the "code wilderness" and build something impactful!
Open to AI/ML or full-stack collaborations globally. Got a challenging idea? Let's build it. π₯