A comprehensive fire safety inspection platform with a Flutter mobile app for field technicians and a Flask API backend. The system manages buildings, fire alarm systems, individual device testing, inspection reports, and service tickets — all with offline-first architecture so technicians can work in basements and mechanical rooms with no cell signal. The Challenge Fire inspection companies deal with a paper-heavy workflow. Technicians visit...

A full-stack e-commerce platform for selling custom canvas art prints. The system features a visual canvas creator where customers design their own prints by selecting backgrounds, stamps, and text overlays, then order them in various sizes and frame options. Includes a complete admin panel, shopping cart, user accounts, and mobile app. The Challenge The client wanted more than a standard Shopify store. They needed a custom canvas design tool...

A sophisticated web scraping system that extracts court case data from county court portals, including filing dates, case numbers, party names, and addresses. The system navigates complex web interfaces protected by AWS WAF and solves audio CAPTCHAs automatically using AI speech recognition. The Challenge Court records are public information, but accessing them at scale is intentionally difficult. The portals use CAPTCHAs, session management,...

An automated stock options alert system that scans the market daily for mean reversion trading opportunities. The system applies a five-condition signal detection algorithm using Bollinger Bands, Keltner Channels, RSI, and ADX indicators, then presents candidates on a Flask dashboard with risk management guardrails. The Challenge Manual stock screening is tedious and emotional. Mean reversion strategies work best when applied consistently...

An embedded hardware system that mounts in a trailer hitch receiver to detect and document dangerous tailgating. The device uses LiDAR for distance measurement, OBD2 for vehicle speed, and a camera module for license plate photography — all coordinated by a Raspberry Pi with intelligent power management. The Challenge Tailgating causes thousands of accidents per year, and victims often have no evidence to support their claims. The goal was to...

A Django-based platform that automates government contract discovery, supplier matching, and outreach for defense procurement. The system scrapes federal databases, extracts award data from PDFs using AI, matches solicitations to qualified suppliers, and manages automated outreach campaigns. The Challenge Government procurement is a massive market, but finding and responding to relevant opportunities is extremely labor-intensive....

A research system — backed by a published paper on arXiv — that enables privacy-conscious LLM inference by splitting transformer models between a local trusted GPU and a remote cloud server. Only intermediate activations (abstract mathematical representations) travel across the network. Raw text never leaves the local machine. The Challenge Organizations in regulated industries (healthcare, legal, defense) want to use powerful language models...

A production AI avatar system that generates real-time lip-synced video of virtual humans speaking generated text. The system combines large language models for conversation, text-to-speech for audio generation, and MuseTalk neural lip-sync to produce video streams delivered via WebRTC with 2–5 second latency. The Challenge Creating realistic, interactive AI avatars that respond in real time is a multi-server orchestration problem. You need...

Developed a Pressure Cycling System Automation solution featuring a simplified web application with real-time data viewing. Created a user-friendly interface with basic password protection,  basic historical data storage/display, and fundamental sensor calibration. Completed server-side development, web application creation, hardware assembly, and testing within a 4-week timeframe. Delivered a streamlined, functional solution meeting the...

Engineered an AI-driven system using Large Language Models and Retrieval-Augmented Generation to transform equipment maintenance processes. The system intelligently processes 100+ page manuals, performs automated maintenance item extraction and database population, and enables dynamic schedule generation. This project drastically reduces manual data entry, improves maintenance efficiency, ensures timely equipment upkeep, centralizes...

I crafted a comprehensive stock evaluation program, harnessing data from the GuruFocus API complemented by targeted web scraping techniques. At its culmination, the tool generates an insightful report, providing a straightforward thumbs up/down verdict on the stock's investment viability. Due to its proprietary nature, further details remain confidential.

Implemented a sophisticated system to monitor pavement thickness, temperature, width, and length in real-time as it's being laid by the paver. Utilizing proprietary sensors and methodologies, this solution ensures meticulous adherence to quality and dimensional specifications, promoting the longevity and consistency of road surfaces.

Designed a comprehensive management software tailored for backflow companies. This tool streamlines workflow processes, tracks equipment status, and oversees technician tasks, ensuring optimized operations and enhanced service delivery in the backflow industry.

Engineered a web scraping utility tailored for attorneys, meticulously sifting through online platforms to identify and collate potential client leads, facilitating targeted outreach and client acquisition for legal professionals.

Crafted a specialized tool to integrate and stream data from an HP Chromatograph using the WITS protocol. This solution ensures seamless data transmission, allowing for real-time analysis and improved operational insights in chromatographic processes.

Developed a web application streamlined for efficient truck and carrier dispatching. The platform facilitates optimized routing, real-time tracking, and seamless communication between dispatchers, drivers, and brokers, ensuring timely and organized operations.

Initiated an ESP32-based solution to oversee an automotive paint booth's functions, including filter cleanliness and various booth components. Integrated with a dedicated web application, this system enables remote, real-time tracking and oversight, ensuring peak operational efficiency and maintenance.

Designed a proprietary software and web application tailored to optimize and simplify the workflows of Home Health Care agencies, driving efficiency across all facets of their operations.

Revamped and modernized Oil/Gas drilling monitoring solution, transitioning its foundation to away from DaqFactory to C#. This refinement optimized the software's performance and scalability while preserving its essential functionalities.

Devised a software that delves into the Handshake blockchain, pinpointing domains nearing expiration. In addition to this core functionality, the tool offers an array of features by leveraging the API furnished by Handshake, enhancing domain management and insights.

Designed a solution to tally items across several conveyor belts, utilizing XBee radios for wireless count transmission to an Arduino Uno collector. This Arduino relayed counts through a cellular modem via HTTP requests to a server. Subsequently, plant stakeholders could access a dedicated web portal, viewing these counts live.

Crafted a unique application harnessing the Twitter API, Wall Street Journal insights, and past stock values to pinpoint potential stock investments. While I'm bound to retain the comprehensive methodology confidential, this robust tool adeptly manages a vast data spectrum.

Leveraging Python and Flask, I engineered a digital platform tailored for vehicle dealerships. The site presented vehicle listings, granting authenticated users access to detailed imagery and specifications. Additionally, users could engage with dealers online, gauge proximity to the dealership, and filter vehicle searches based on desired criteria. Reporting functionalities were also integrated for enhanced utility.

Devised a SCADA-powered interface for evaluating electrical submersible pumps. Beyond core machine operations, the software interconnected with a MySQL database I architected, housing current pump tests alongside manufacturer's optimal efficiencies. Through a graphical interface, the software juxtaposed real-time pressure and power curves against standard curves, offering precise pump efficiency insights.

Crafted a software suite to actively track metrics like depth, lag depth, and pump activities, inclusive of three distinct pumps. The system also managed gas measurements, chromatography, and on/off bottom statuses. Integrated with WITS Level 0 standards, the software not only facilitated live operations with dynamic graphing features but also liaised with other onsite systems, sourcing data from instruments I personally designed.