Skip to content
View aryanyaksh-art's full-sized avatar
1 follower · 0 following

Block or report aryanyaksh-art

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
aryanyaksh-art/README.md

header

Hi there, I'm Aryan Yaksh 🚀

typing

📍 Greater Toronto Area | 🛰️ Independent researcher and builder

I build the software that gets spacecraft down safely and check it against what NASA and ESA actually measured. Right now I'm rebuilding OSIRIS-REx's autonomous landing system from scratch (hazard-detection AI, navigation and a replay site) and reproducing Planck's cosmic microwave background results with the CLASS code. Space is the mission, and AI is the toolkit I use to get there.

🛰️ Featured Projects

🌑 touch_down

Rebuilding OSIRIS-REx's autonomous landing system in simulation, with my own dataset, hazard-detection AI and navigation, then validating against NASA's published results.

🌌 cmb-peak-height-cdm-sensitivity

Reproducing the CMB acoustic-peak and dark-matter-density relationship with the CLASS Boltzmann code, validated against Planck 2018.

🤖 did-it-even-learn

Do community LLM fine-tunes beat their base model, or just memorize the benchmark? Contamination-controlled study across hundreds of matched pairs.

🧬 drug-match

Finds already-approved drugs that could treat diseases with no good treatment, and shows the drug → gene → disease path behind each suggestion.

📡 Currently Working On

  • 🌑 Validating a simulated OSIRIS-REx landing against NASA's published results
  • 🌌 Checking cosmology results against real Planck 2018 data
  • 🤖 Finding out which LLM fine-tunes really learn something new

🧰 Tech Stack

tech stack

📨 Let's Connect!

I'm always open to collaborating on interesting projects or talking space and research.

Instagram LinkedIn

contribution snake

footer

Pinned Loading

  1. cmb-peak-height-cdm-sensitivity cmb-peak-height-cdm-sensitivity Public

    Reproducing the CMB acoustic peak-height dependence on the cold dark matter density with the CLASS Boltzmann code, validated against Planck 2018.

    Jupyter Notebook

  2. did-it-even-learn did-it-even-learn Public

    Do community LLM fine-tunes actually improve over their base, or just memorize the benchmark? A contamination-controlled study across hundreds of matched Featherless-hosted pairs.

    HTML

  3. levelup-reader levelup-reader Public

    TypeScript

  4. volleyball-simulator volleyball-simulator Public

    3D volleyball strategy simulator for coaches. Plan rotations, script plays, and check serve-receive/attack-defense coverage.

    TypeScript