Headshot of Jack P. DeMarinis

Jack P. DeMarinis

About Me

Computer engineer with experience in Generative AI and ML, robotics software, autonomy-adjacent systems, and simulation. Experienced building multi-agent robotic systems using ROS on Linux, integrating real-world sensor data, and developing performance-aware C++ and Python software. Background spans robotics middleware, embedded Linux development, simulation, and AI-driven decision systems. Comfortable working across software and hardware boundaries in test-heavy, iterative environments.

Education

Master of Science: Electrical Engineering

University of Rhode Island - Expected May 2026

Accelerated B.S./M.S. (ABM) Program

GPA: 4.00 / 4.00

Bachelor of Science: Computer Engineering

University of Rhode Island - Graduated May 2025

Minor: Mathematics

GPA: 3.90 / 4.00

Technical Skills

Languages

  • C++
  • C
  • Python
  • Bash
  • JavaScript
  • HTML/CSS
  • MIPS Assembly
  • LC-3

Robotics & Systems

  • ROS / ROS2
  • Linux (Ubuntu)
  • Docker
  • Robotics middleware
  • Simulation pipelines
  • Sensor integration (odometry, RGB-D)
  • Real-time data processing
  • SWARM

AI / Autonomy Tooling

  • PyTorch
  • Hugging Face Transformers
  • Retrieval-augmented generation (RAG)
  • Embeddings
  • FAISS vector search
  • LLM-based decision systems
  • GCP

Engineering Tools

  • Unity (3D simulation & visualization)
  • MATLAB
  • Fusion 360
  • AutoCAD
  • LTSpice
  • Multisim
  • VHDL
  • OpenMV IDE

Experience

Graduate Research Assistant

University of Rhode Island

  • Worked on SWARM simulation through Unity using decision-making pipelines, emphasizing reliability, observability, and fault handling.
  • Developed VR simulation environments to exercise system behavior.
  • Implemented reproducible Docker-based workflows to support consistent builds, testing, and deployments across machines.

Undergraduate Research Assistant

University of Rhode Island

  • Built and evaluated backend software components supporting LLM decision workflows.
  • Developed and tested custom RAG pipelines for context management using FAISS and BM25-based retrieval (vector and keyword search).
  • Created evaluation scripts to compare LLM variants using consistent test sets and measurable latency, cost, and output-quality criteria.

Computer Engineering Intern

Electro Standards Laboratories

  • Developed and maintained embedded Linux software using C and Python.
  • Troubleshot electrical and software issues through hands-on testing and system-level debugging.
  • Collaborated with hardware and engineering teams to validate system behavior and reliability.
  • Authored technical documentation supporting long-term maintainability and deployment.

Software Engineering Intern

IGT

  • Debugged and resolved Linux system issues using low-level command-line and system analysis tools.
  • Developed Bash, C, and C++ software for device-level API integration.
  • Implemented and evaluated OCR pipelines to improve system accuracy and robustness.

Projects

AI Meeting Assistant

Developed an end-to-end meeting intelligence platform including transcription and summarization. Integrated speech-to-text systems with LLM-based action item and agenda extraction, emphasizing modular design and reliability across varying input conditions.

Visit Meeting Muncher

Senior Capstone: Robotic Assembly & Inspection

Designed an automated PCBA assembly and inspection workstation. Integrated robotics hardware, sensors, and control software into a unified system, and fabricated custom parts using 3D printing.

View Project Report

Agentic RAG Chatbot

Built a multi-agent retrieval-augmented generation system for structured reasoning. Implemented embedding pipelines and FAISS-based vector search to reduce hallucinations and improve response quality. Deployed as a Dockerized Flask API designed for iterative testing.

Try the Chatbot

LLM Robotics

Built and simulated a TurtleBot-style Waffle robot using ROS on Linux, integrating odometry and RGB-D camera data into a unified perception pipeline. Designed ROS nodes to publish structured state representations and connected real-time observations to a large language model for high-level reasoning and decision-making.

Honors & Activities

Contact

Open to new opportunities and collaboration. Reach out anytime.