Skip to content
View ryanmccormickdev's full-sized avatar

Highlights

  • Pro

Block or report ryanmccormickdev

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ryanmccormickdev/README.md

Ryan McCormick

Machine Learning Engineer @ Bedrock ResearchCS @ Stanford
I build geospatial ML systems: foundation-model embeddings, unsupervised clustering, anomaly/rare-object detection, and tooling for production-ready applications.

GitHub · Email · LinkedIn


What I’m best with

  • Cross-modal Foundation Models — embeddings across optical, SAR, and thermal; unified latent spaces; zero/low-label workflows.
  • Unsupervised Tooling — HDBSCAN pipelines, cluster-quality metrics, radar-plot diagnostics, CLI for large-scale embedding jobs.
  • Anomaly & Rare-Object Detection — sliding-window + target-similarity scoring; fast iteration on large overhead datasets.

“From all of us at Bedrock: best of luck, Ryan… you’ve set the bar high.” — Matt Reisman, CTO, Bedrock Research


Selected Work

  • Latent Spaces: Seeing Inside an AI Brain — Blog post and demos on unsupervised geospatial modeling
    https://www.bedrockresearch.ai/latent-spaces-seeing-inside-an-ai-brain
  • (Soon to be published 2nd blogpost, not public yet)
  • Embedding & Clustering Suite (Bedrock) — modular metrics (incl. density-aware), visualizations, CLI, containerized.
  • Anomaly Detection Pipeline (Bedrock) — unsupervised scoring + human-in-the-loop exploration.
  • Metadata Search for Massive Datasets — SQL schema + fast geo filtering (country/lat-lon), Overture + Google Maps integration.
  • I have my name as a co-author on a few papers (see my resume for a detailed list), but I have yet to write my own study independently. I wouldn't be surprised if that changes by the end of this year.

Public Projects

  • Self-Balancing Robot (WIP) — RL + mechatronics sandbox
    Repo: https://github.com/Giantryan484/Balancing-Robot
  • TabGenerator.app — audio → playable guitar tabs
    Repo: https://github.com/Giantryan484/Guitar-TAB-Generator · Live: https://tabgenerator.app/
  • Mini-Projects — small tools and experiments (lots of tiny wins)
    Repo: https://github.com/Giantryan484/Mini-Projects
  • Frisbee Drone (Big WIP) - Not ready to be shown off quite yet, but I'm intermittently working on a CV system for tracking thrown frisbees and retrieving them with a drone.
    Repo: None, yet
  • Personal Website - ryanmccormick.dev (Soon to be published) - I'm going a bit over-the-top with my website design to practice my frontend skills. So, there's a lot of assets I need to make beyond my traditional programming tasks. It could be a while before I get this looking good.
    Repo: None, yet

Many of these are from my time in high school and have been abandoned in states of ~60% functionality because I had such an experimentation-oriented brain at the time. Please give me a chance to prove my coding abilities beyond this mess.


Tech I use

Python, PyTorch, TensorFlow/Keras • Docker, AWS, Google Cloud • HDBSCAN, scikit-learn, numpy/pandas • Raster/SAR tooling (sarpy, tif stacks) Full-stack when needed: JS/HTML/CSS, React/SvelteKit, SQL · Git, Bitbucket, CI/CD

Certs: TensorFlow Professional Developer · SolidWorks Expert (top <1%)

I'm also very familiar with C++ and Java through the courses I've taken here at Stanford. I just haven't found use for them in any of the projects I've worked on yet.


Highlights

  • Built & shipped unsupervised geospatial ML tooling used in client-facing deliverables.
  • Designed containerized CLIs to abstract away the complexity of cluster visualization and analysis.
  • Created anomaly & rare-object detectors that performed across SAR/optical without labels.
  • Wrote externally-facing technical content and presented to government and industry audiences.

A Note About My Future

Right now I'm focusing on my studies at Stanford. Don't expect regular updates to any of these projects or documentation; I'm only able to work in between problem sets and over breaks. My coursework is quite comprehensive, covering everything from standard DSA to more detailed courses in fully-autonomous robotics and safe + reliable AI systems. I'm very early in my career and I'm prioritizing a strong base and foundation in AI research before I dive into industry head-on. So, even though I might not have as flashy products as others, you can be sure I'm one of the hardest workers and quite a breeze to solve the world's toughest problems beside. If Stanford let me in for that merit, you can be certain your company will love it just as much.

If you think I'd be a good fit for one of your projects, tell me now! As a freshman I'm flying under the radar of most companies. At the risk of being egotistical, you should get in touch with me before I'm a senior engineer you'll have to pay 10x as much.

Contact me at hi@ryanmccormick.dev

Pinned Loading

  1. Mini-Projects Mini-Projects Public

    This repo documents my small bursts of engineering design throughout my time in high school. Read the documentation to learn more about my motivations as an engineer!

  2. Guitar-TAB-Generator Guitar-TAB-Generator Public

    This repo documents the first website I've built with a custom AI Integration. Read the documentation and try out the website (tabgenerator.app) to learn more about this interdisciplinary project!

    JavaScript

  3. Balancing-Robot Balancing-Robot Public

    This repo documents my first robotics project, SADIE, which is quite a complex one. Read the documentation to learn about my exploratory learning process and my approach to systems engineering!