Skip to content
View Mkhurana3's full-sized avatar
🎯
Deep Sleep
🎯
Deep Sleep

Block or report Mkhurana3

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. 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
Mkhurana3/README.md

Pixel-art ML lab: CRT monitor drawing a loss curve, neural network, server racks

✦ Manan Khurana


Mkhurana3 · CS undergraduate · machine learning · data science · computational biology

Machine learning for biological data

Animated DNA sequencer translating a gene


✦ about me

; manan.asm · x86-64 · NASM

section .data
    name      db "Manan Khurana", 0
    handle    db "Mkhurana3", 0
    location  db "Delhi, India", 0

    focus     db "Machine Learning", 0
              db "Data Science", 0
              db "Computational Biology", 0

    open_to   db "Research internships", 0
              db "Open-source collaboration", 0

section .text
    global _start

_start:
    lea   rdi, [rel focus]
.loop:
    call  read_papers
    call  write_code
    call  ship
    jmp   .loop             ; never returns
Python session training a neural network

✦ tech stack

Languages

Machine Learning & Data Science

Bioinformatics

Web

Tools & Platforms

Animated neural network: forward pass and backpropagation


✦ featured projects

AushadhiTrack start page: the latest hash-chained ledger records and a check that every record still verifies

💊 AushadhiTrack

Smart India Hackathon 2026 prototype (SIH26198) that follows Jan Aushadhi medicine from warehouse to patient. A SHA-256 hash-chained ledger, seven explainable fraud rules, an Isolation Forest that catches patterns the rules miss, and Ed25519-signed QR receipts patients can check in English or Hindi.



Score of the real genetic code against 100,000 random codes

🧬 codon-shuffle

Is the genetic code better than random? A reproduction of Haig & Hurst (1991): we scored 100,000 random genetic codes on how much damage a single mutation does, and only 10 beat the real one. Plus an early AlphaFold follow-up on folded vs disordered protein regions. Built with Antara: I led the biology and worked on the code with her.




✦ github stats

GitHub activity: contributions, streaks, heatmap and weekly trend

auto-updated daily by a GitHub Action


✦ find me

Pixel-art lab ending screen: microscope, DNA helix, sequencing readout — thanks for playing

Popular repositories Loading

  1. AushadhiTrack AushadhiTrack Public

    TypeScript

  2. ocarinaforyou ocarinaforyou Public

    Forked from seedds/nxnative

    Ocarina of Time on a MacBook Air M2 with nxnative: 1080p settings, notes on the Saria cutscene GPU hang, and test scripts

    Python

  3. Mkhurana3 Mkhurana3 Public

    Python

  4. codon-shuffle codon-shuffle Public

    Is the genetic code better than random? 100,000 fake codes + an AlphaFold follow-up

    Jupyter Notebook