|

I am Aaryan Tahir, Electrical & Computer Engineer • AI & Systems Builder. 

I build and deploy production machine learning systems, real-time edge inference pipelines, and scalable on-premise LLM platforms — bridging deep hardware engineering with applied AI.

Scroll to Explore
About

I engineer ideas into reliable, real-world systems. 

I am an Electrical and Computer Engineering student at Drexel University concentrating in Computer Science, with enterprise AI co-op experience at Susquehanna International Group (SIG) and Exelon/PECO. I build and deploy production machine learning systems, real-time edge inference pipelines, and scalable on-premise LLM platforms.

As Lead Researcher and Founder of Drexel Alumni Labs (Vision AI), I lead a team of 5 researchers developing edge-inference computer vision devices that monitor 40+ camera feeds simultaneously on Nvidia Jetson hardware. My work bridges deep hardware engineering, autonomous agent architectures, and enterprise AI systems.

About me
Experience

Career 

Susquehanna International Group (SIG)

Artificial Intelligence Co-op

Building LLM- and NLP-powered automation tools at a leading proprietary trading firm. Developing production-oriented AI pipelines across cloud infrastructure (AWS/Azure/GCP), applying modern MLOps practices.

Exelon – PECO | Technical Services

AI & Automation Intern

Deployed PM-AI, an enterprise RAG-powered AI Chatbot trained on 15+ years of data. Engineered AVIS, a fully autonomous LLM + computer vision defect detection system with companion drone automation.

Exelon – PECO | Distribution Standards

Preventative Maintenance Engineer Intern

Deployed a Department Server to host automated scripts and Power BI dashboards. Automated monthly PUC reporting and Circuit Patrol, saving 20+ hours monthly.

Drexel University | College of Engineering

Lead Researcher & Founder: Vision AI

Led a team of 5 researchers deploying an edge-inference computer vision safety device on Nvidia Jetson hardware. Achieved 94% detection accuracy and 73% reduction in false alarms.

Drexel University | College of Computing

Senior Technical Ambassador

Directed a team of 12 Technical Ambassadors across Canvas, Blackboard, and 20+ hackathons. Engineered an LLM-powered ticket automation system.

Education

B.S. Electrical & Computer Engineering (CS Concentration) — Drexel University • Dean's List • Founder Scholar • STAR Scholar

Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 • Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 • Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 •

Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 • Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 • Python • PyTorch • LangChain • RAG • OpenCV • HuggingFace • C/C++ • React • Next.js • SQL • Docker • Linux • AWS • Nvidia Jetson • Arduino • ESP32 •

Selected Projects

Work 

  • Vision AI (Drexel Alumni Labs)

    Edge AI & Threat Detection

    Autonomous edge-inference computer vision safety platform monitoring 40+ camera feeds simultaneously.
  • PM-AI & AVIS Defect Detection

    Enterprise GenAI • Exelon/PECO

    Enterprise RAG-powered chatbot trained on 15+ years of departmental data for 100+ active users.
  • CCI-Bot (Drexel CCI)

    On-Premise Enterprise LLM

    On-premise LLM chatbot trained on 10+ years of Drexel support data, serving 5,000+ students monthly.
  • Aarz

    AI Agent Architecture

    Autonomous agent runtime with an adaptive planning loop, local tool-calling protocol, and custom memory.
  • Drexel Bench

    IoT & Smart Campus

    IoT seating tracker using Raspberry Pi/ESP32, ultrasonic sensors, OLED displays, and a web portal.
  • What's in My Fridge

    Full Stack Web App

    Interactive recipe suggestion web app that dynamically generates meal ideas from available pantry ingredients.
View