Delivering end-to-end full stack and AI-powered solutions for a private client, building a portfolio of high-impact products across fintech, real estate, and productivity domains.

Built an autonomous trading strategy generator and tester powered by the Claude API, enabling automated generation, backtesting, and evaluation of trading strategies without manual intervention.

Developed multiple market making bots for automated order management and liquidity provision on cryptocurrency exchanges.

Created a Hyperliquid data collection and analytics tool that crawls on-chain trading data to identify and rank top-performing traders, providing actionable market intelligence.

Developed a Brand Performance SaaS application that benchmarks how brands are perceived and ranked across leading AI model responses — offering businesses a novel way to monitor and improve their AI-era brand visibility.

Built a Python Debugger tool powered by the Claude API that intelligently analyzes, explains, and resolves code errors, streamlining the debugging workflow for developers.

Designed and developed a real estate platform allowing users to browse, rent, and purchase properties, featuring full property listing management and user flows.

Engineered a feature-rich iOS Personal Assistant app with deep automation capabilities including meeting scheduling, music playback, calls, email reading and sending, AI image generation and editing, and a social layer with shared friend activities and personal/group chat.

Contributed to the development of an AI-powered accounting platform, combining full-stack engineering and artificial intelligence to enhance automation and efficiency.

Developed and deployed an AI-powered accounting software using Python (Flask) for backend services and Supabase for database management and authentication.

Designed and integrated AI agents to automate data processing, enhance accuracy, and streamline financial operations.

Built and optimized RESTful APIs for seamless communication between frontend and backend systems, improving performance and scalability.

Collaborated with cross-functional teams to deliver production-grade, secure, and user-friendly financial solutions within project deadlines.

Developed and deployed machine learning models for financial price forecasting using OHLCV data enriched with technical indicators. Experimented with and benchmarked a wide range of models including ANN, ARIMA, Prophet, Linear Regression, and ensemble methods to optimize prediction accuracy. Built and exposed model inference via Flask-based REST APIs for real-time integration. Continuously iterated on feature engineering pipelines to improve signal quality from raw market data.

Designed and built an autonomous trading bot system capable of running multiple bots in parallel on a live trading platform. Implemented a dry-run performance evaluation mechanism that automatically benchmarks active models and dynamically switches to the best-performing one — reducing manual intervention and improving trading strategy adaptability in real market conditions.

Conducted applied AI research at the National Center for Cyber Security, focusing on two core areas. First, image forgery detection — developing and evaluating deep learning models to identify manipulated or tampered digital images, contributing to digital forensics capabilities. Second, harmful object detection and image analysis — building computer vision pipelines to detect dangerous or suspicious objects in visual data for security applications. Worked across the full ML lifecycle from data collection and preprocessing to model training, evaluation, and reporting.

Contributed to applied AI and digital forensics research at the National Center for Cyber Security. Played a key role in the full-stack development of a mobile forensic application — designing and building end-to-end features for extracting, analyzing, and presenting digital evidence from mobile devices. Collaborated with the research team to integrate AI-driven analysis capabilities into the forensic pipeline, supporting real-world cybersecurity investigation workflows.