I'm Tyrone Marhguy
I'm Tyrone Marhguy
I'm Tyrone Marhguy
Building everything from 700+ discrete-transistor CPUs to scalable web platforms serving 350K+ users. Computer Engineering student at UPenn passionate about bridging silicon and software to create impactful solutions.
My name is Tyrone.
I design discrete-transistor processors and build AI-powered applications that solve real-world problems. From hardware architecture to full-stack development, I bridge the gap between silicon and software.
Want to see my work? Explore my projects, learn about my journey, download my , or connect with me on LinkedIn.
Flagship Projects
From hardware design to machine learning, these projects showcase the breadth and depth of my technical expertise.
8-bit Transistor CPU
Complete CPU built from 700+ discrete 2N7000 NMOS transistors with custom 12-instruction ISA, capable of running assembly programs and measured with logic analyzer for gate propagation delays.
MoMo Credit Score Generator - Ghana
Ghana's first consumer credit platform processing 1M+ transactions with FastAPI microservices and XGBoost ML pipeline, achieving AUC 0.82 and improving lender approval rates by 32%.
Music & You
ML-based music analytics platform (FastAPI, OAuth2) supporting 1M+ users with <2s P95 latency, featuring React interface with async streaming that reduced load times by 40%.
UniBridge - Ghana
Ghana's first centralized admissions portal for 350K+ applicants/year using FastAPI, React, Redis, and PostgreSQL, achieving 99.9% uptime and P99 <500ms latency under 10K+ daily requests.
Software Development
From psychology-based games to automation tools, exploring diverse applications of software engineering.
UniBridge-Ghana
Full-stack student platforms (NextJS, Express, MongoDB) handling 800+ MAU with automated scheduling, real-time chat, and SPA with multi-course management.
MoMo Credit Score Generator - Ghana
Ghana's first consumer credit platform processing 1M+ transactions with FastAPI microservices and XGBoost ML pipeline, achieving AUC 0.82 and improving lender approval rates by 32%.
Music & You
ML-based music analytics platform (FastAPI, OAuth2) supporting 1M+ users with <2s P95 latency, featuring React interface with async streaming that reduced load times by 40%.
AtomAssembler
Two-pass assembler translating 20+ mnemonics to Intel-HEX for custom 8-bit CPU, reducing firmware build time by 70% with automated testbench validating 50+ programs at 100% pass rate.
Color Communication Game
Multiplayer psychology-based game where users develop shared language using color patterns, exploring behavioral communication and emergent semantics.
YouTube to Spotify Migrator
CLI tool for seamless playlist and liked songs migration from YouTube Music to Spotify using OAuth and API automation.
Hardware Engineering
Building from transistors up, creating physical computing systems and culturally-inspired automation.
8-bit Transistor CPU
Complete CPU built from 700+ discrete 2N7000 NMOS transistors with custom 12-instruction ISA, capable of running assembly programs and measured with logic analyzer for gate propagation delays.
16-Bit Transistor Memory Module
16-bit addressable memory from ~350 discrete transistors using D-flip-flop latch arrays, integrated with Arduino microcontroller achieving stable operation over 1,000+ access cycles.
Precision Clock Oscillator
RC multivibrator delivering 1 kHz clock with <1% drift, tuned for stable duty cycle and temperature resilience using discrete components.
Transistor-Logic Calculator
Four-function calculator using ripple-carry adders and custom 7-segment decoder with power management, achieving <5ns propagation latency for all 16-bit operations.
Technical Skills
Core technologies and tools I work with across hardware design, software development, and data science.
Programming
Web & APIs
Data & Cloud
Hardware Design
DevOps & Testing
ML & Analytics
Currently Learning
Byte-Sized Bio
I use transistors and algorithms to build cool stuff that empowers people to create.
Building the future through hardware innovation and principled problem-solving
Engineering Philosophy
I believe in building from first principles—whether designing transistor-level CPUs or developing ML algorithms. My approach combines rigorous technical execution with a deep understanding of real-world constraints and user needs.
Core Competencies
Resilience & Advocacy
My advocacy for student rights in Ghana gained international recognition [Wikipedia ↗], while simultaneously achieving top 0.1% in WASSCE. This experience taught me persistence and principled problem-solving—values I now apply to transistor-level CPUs and privacy-centric ML.
Timeline Highlights
📄Download My Resume!
Computer Engineering student at UPenn with specialized resumes for hardware engineering (discrete systems, FPGA, embedded) and software development (full-stack, ML, fintech) roles.