Golang Developer

Integrated real-time machine learning models into fraud detection pipelines using Go, Python, and Tecton, decreasing fraud losses by 10% and false positives by 20%, saving $2M annually across 100M+ transactions. Built a React + TypeScript-based spam detection quiz across web and React Native mobile platforms using Next.js and ReactDOMServer; increased user verification compliance by 25%, preventing $1.5M in fraud. Migrated rule engine from Ruby on Rails to Go, enhancing transaction validation with Go’s concurrency and static typing to improve processing speeds. Implemented configurable send/withdrawal delays in Go. Integrated cross services with Kafka and DynamoDB, reducing risky withdrawals by 12% and false positives by 18% across $100M+ in transactions.

Spearheaded designed and developed an Advanced Bug Reporting Tool; engineered with React.js, Tailwind CSS, Material UI, JavaScript, TypeScript for frontend and GraphQL, and Node.js for backend interactions. Boosted entire user feedback precision by 50% by implementing advanced features like screenshot commenting, drawing utilities, and video capture to boost user interaction and significantly optimize the bug-tracking.

Trained and fine-tuned XGBoost machine learning model, resulting in an accuracy of 80.8% for stress level classification prediction from multimodal physiological signals, outperforming previous studies using Python, PyTorch, and scikit-learn.