Say Hello
I'm an Electrical & Computer Engineer who builds and ships full‑stack AI products. From embedded systems and hardware design at Rutgers, to founding and engineering multiple live SaaS platforms — I turn complex technical problems into elegant, revenue‑generating products.
Work Experience
Sole engineer on ClawbackAI — a production SaaS platform I architected and shipped from scratch for Hummus Republic. The platform is actively used by 40 franchise locations and 100+ franchisees, replacing disconnected spreadsheets and email chains with a unified, real-time operations hub.
The Problem
Franchise leadership had zero visibility into per-unit financial performance or compliance status. P&L data lived in disconnected spreadsheets, SOPs were emailed PDFs, and billing reconciliation was done manually — causing missed clawbacks worth thousands of dollars per quarter.
The Solution
Architected a multi-tenant React + Firebase platform with 4-tier RBAC, a Google BigQuery analytics layer aggregating financials across all franchise units, and an AI billing reconciliation engine that automatically surfaces clawback discrepancies.
Key Accomplishments
- ▸4-tier RBAC system — corp admin, franchise owner, location manager, and employee access levels
- ▸Google BigQuery pipeline aggregating P&L, COGS, and ops data across all franchise locations in real time
- ▸AI billing reconciliation — auto-detects clawback gaps that previously required hours of manual review
- ▸Document management with versioned SOP knowledge base — reduced franchise support requests significantly
- ▸Construction tracker — manages bid submissions, stage budgets, and contractor dashboards for new locations
- ▸Real estate acquisition pipeline for new franchise location prospecting and deal tracking
- ▸Firebase Cloud Functions powering automated report generation, compliance alerts, and scheduled syncs
- ▸Sole engineer: designed schema, built all features, deployed, and maintain in production
Tech Stack
Founded and built BriefRoom — a live AI interview prep platform serving job seekers across 440+ companies and 6 interview formats. Designed, engineered, and shipped the entire product solo, from database schema to Stripe billing to AI prompt architecture.
The Problem
Generic interview prep tools don't simulate real company formats or give meaningful feedback. A candidate prepping for Google system design needs a fundamentally different experience than one prepping for McKinsey case interviews — no existing tool provided that specificity.
The Solution
Built a multi-model AI system with company-specific interviewer personas calibrated to real hiring patterns at 440+ companies. Each session delivers STAR-scored feedback, cites the candidate's exact words, and tracks improvement across sessions.
Key Accomplishments
- ▸440+ company-specific AI interviewers — behavioral, coding, system design, case, medical, and law formats
- ▸4.8★ aggregate rating across 50+ verified user reviews on the live platform
- ▸Multi-model AI architecture — different models optimized per interview type for response quality
- ▸STAR scoring engine — AI quotes exact candidate answers with structured improvement feedback
- ▸4-tier Stripe billing: Free → $9/wk → $19/mo → $49 Career Launch — full subscription lifecycle
- ▸Session history and improvement analytics — users track score trajectory across practice rounds
- ▸Built and shipped solo: architecture, AI prompting, billing, auth, and deployment
Tech Stack
Projects
Capstone Design Project
BreezeBeam
Senior capstone design project integrating embedded hardware and software for intelligent environmental monitoring and control. Designed custom PCB circuitry, wrote embedded firmware, and built a software interface for real-time control.
- ▸Custom PCB design with ECAD simulation and LTSpice validation
- ▸Embedded C/C++ firmware for sensor data acquisition and control logic
- ▸Hardware-software interface for real-time environmental parameter control
Embedded Systems
Wi-Fi Signal Meter
ESP32-based device that scans nearby Wi-Fi networks and visualizes signal strength in real time using an RGB LED bar and LCD display. Signal bands mapped to color gradients for intuitive UX without a phone app.
- ▸ESP32 firmware scanning RSSI values across all visible SSIDs
- ▸RGB LED array with PWM-driven color gradients (green → red) per signal band
- ▸16×2 LCD display with dynamic scrolling network list and RSSI readout
Smart Lighting System
LumenOS
Intelligent ambient lighting system that automatically adjusts color temperature and brightness based on occupancy, ambient light, and temperature. IR remote control, multi-zone support, and a web dashboard for manual override.
- ▸PIR motion sensor + LDR photoresistor fusion for adaptive auto-on/off
- ▸DHT11 temperature sensor adjusts warm/cool bias for circadian rhythm support
- ▸IR remote with 20+ programmable presets; all state persisted in EEPROM
AI Projects · Embedded Systems
AI & Embedded Research
Deep learning models, image recognition systems, and embedded AI implementations from Rutgers ECE coursework. CNNs, FPGA accelerators, and classifier comparisons across benchmark datasets.
- ▸ResNet50 image classification — fine-tuned on custom datasets, benchmarked against baselines
- ▸AlexNet CIFAR-10 optimizer study — 48 experiments across Adam/SGD/RMSProp variants
- ▸Naive Bayes vs. Perceptron — 76% accuracy on digits, 85% on face recognition datasets
Education
Bachelor of Electrical And Computer Engineering
Rutgers University
Developed critical problem-solving and collaboration skills through fast-paced, technically rigorous coursework. Focused on hardware-software integration, embedded systems, and AI applications.
Minor in Computer Science
Rutgers University
Focused on artificial intelligence and cybersecurity — developing skills in machine learning, secure coding, and applying AI to real-world problems.
Skills
Let's Build Something
Open to roles, collaborations, and interesting problems.
Let's Get in Touch!
Whether you're a recruiter, potential co-founder, or just want to talk tech — my inbox is open.