Reed Pennock

I build reliable software for scientific computing, data pipelines, and intelligent developer workflows.

about

I am a B.S. Computer Science student at Brigham Young University in Provo, Utah, graduating in April 2028. I work across scientific software, compilers, computer vision, and web applications.

At OpenTeams, I work on statistical computing, compiler correctness, and developer workflows. I also build neuroscience data pipelines at BYU and lead Linux and open-source workshops for the BYU Linux Association.

experience

Software Engineering Intern, Product Team

OpenTeams

Built ppstats, a 27-kernel statistics package; native skew and kurtosis ran about 25 times faster than SciPy on arrays of one million elements.

  • Submitted PostPython compiler fixes for silent Float16 miscompilation, non-finite constants, and C symbol collisions, with end-to-end regression tests.
  • Contributed QA and UI/UX fixes to Collab, an AI workspace; led release-note videos and documentation and built Claude Code and Codex workflows for implementation and code review.

Software Developer, Neuroscience Lab

Brigham Young University

Leading development of Python pipelines that process terabytes of calcium-imaging video and EEG/electrophysiology data for a nine-person neuroscience research team.

  • Re-architected the analysis package into extensible object-oriented components and introduced GitHub Actions CI and comprehensive end-to-end testing.

Undergraduate Teaching Assistant, CS 111

Brigham Young University

Mentored more than 100 students in Python, recursion, object-oriented programming, and software testing through labs, office hours, and individual code reviews.

  • Reviewed more than 200 programming projects and provided actionable feedback on correctness, algorithmic efficiency, code quality, and style.

projects

Gym Occupancy Analytics Pipeline

Python, PyTorch, YOLO, ONNX Runtime, SQLite, Linux

Built a fault-tolerant pipeline that collects, deduplicates, and stores timestamped campus gym imagery and metadata after reverse-engineering an Android camera API and anonymous OAuth flow.

  • Developed a computer-vision pipeline using YOLO, tiled inference, spatial zones, and equipment polygons; reduced mean count error from 6.0 to 1.5 people in an initial manually labeled evaluation.
  • Deployed collection and inference on Linux with systemd recovery, stale-feed detection, atomic database snapshots, and secure Tailscale/Syncthing synchronization; optimized GTX 1070 Ti inference to run nine detection passes per image in about 0.5 seconds.

Easy Grocer

Next.js, TypeScript, Supabase, PostgreSQL, Tailwind CSS

Built a progressive web app that generates personalized weekly meal plans from dietary, budget, calorie, and meal-pattern preferences.

  • Implemented Supabase authentication, persistent planning history, offline support, database migrations, cached product resolution, Walmart cart handoff, and automated CI and security checks.

Chess Web Server

Java, WebSockets, SQL

Built a real-time multiplayer chess platform with concurrent game sessions, legal-move validation, authentication, and persistent game storage.

ppstats

Python, PostPython, NumPy, SciPy

Built a 27-kernel statistics package in PostPython and cross-validated Python, native shared-library, and NumPy ufunc results against SciPy and NumPy.

  • Native skew and kurtosis ran about 25 times faster than SciPy on one-million-element arrays in benchmark runs.

ACE Neuro

Python, calcium imaging, electrophysiology

Developed a modular analysis pipeline for calcium-imaging video and EEG/electrophysiology data, with components for processing, visualization, and aligning recordings across modalities.

  • Reworked the package into extensible components and added CI and end-to-end tests for the neuroscience research team.

Finnish Inflection CLI

Python, UralicNLP, Omorfi

Built an interactive command-line tool for practicing Finnish noun, adjective, and verb inflections with morphology-generated answers and focused drills.

  • Added custom vocabulary imports with validation, review of ambiguous forms, and safe replacement of saved sets.

Suomi Scraper

Python, Voikko, Streamlit

Built a tool that extracts Finnish vocabulary from web pages and documents, reduces words to their base forms, and exports reviewable Anki-ready decks.

  • Supports PDF, DOCX, and text input, cached translations, and a browser interface for editing cards before export.

education

Brigham Young University

B.S. in Computer Science

  • Vice President, BYU Linux Association (2026–present): lead workshops for 20–30 attendees on Linux, open-source software, and systems tools; help students make open-source contributions.

technology stack

languages

  • Python
  • C
  • C++
  • Java
  • SQL
  • TypeScript
  • JavaScript
  • HTML/CSS

frameworks & libraries

  • React
  • Next.js
  • FastAPI
  • pandas
  • NumPy
  • SciPy
  • Matplotlib
  • scikit-learn
  • pytest
  • PyTorch
  • ONNX Runtime

tools & platforms

  • Linux
  • Git
  • GitHub Actions
  • Docker
  • AWS
  • PostgreSQL
  • Supabase
  • Redis
  • Celery
  • systemd
  • Tailscale

AI-native development

  • Claude Code
  • OpenAI Codex
  • Agent orchestration
  • AI-assisted development
  • QA
  • Technical documentation

human language

  • Finnish

Let's build something useful.

For software, research, or open-source collaboration, reach out by email or find me on GitHub and LinkedIn.

reederpen@gmail.com