Implemented Simple Linear Regression from scratch in Rust (OLS).
Computing the coefficients b0, b1 directly from the formula.
Rust’s map and iter pipelines took some getting used to really a big pain, but it was a fun exercise.
Will optimize it and keep building more ML implementations.
Spent last night deep in Rust!
Implemented Random Forest on the Iris dataset
will soon try some other dataset, with manual preprocessing
Got the accuracy of as: 97.78%
In Rust you feel every single line
Been learning Rust for a few days now.
Started with the basics, structs, enums, Ownership, Mutability, Ownership, Borrowing, References etc .
Curiosity hit can I actually do ML in Rust?
Found this from @freeCodeCamp
[Link:- https://t.co/J4MYur4jxM]
linfa makes it possible.
Played with Decision Tree on iris:
-> Gini, max_depth=3, 85% split → 86.36% accuracy
-> Entropy, max_depth=10, 85% split → 81.82%
Smaller training data + deeper tree = overfitting. Classic.
Well today office was empty
So spent the time exploring MLflow properly for the first time on a small language detection work.
Nothing fancy, just wanted to see how experiment tracking actually works outside of tutorials.
Link - https://t.co/ESm3xbbHqi
busy with the internship but yeah
finished building voice translator that runs fully local, no API costs😁
upload or record audio → transcribe → translate
Hindi, Arabic, Spanish, French, English :)
link:- https://t.co/eFYkT2PZPW
-> Started building voice message translator today
-> Whisper STT base model is working now
-> Output isn't great yet but the pipeline works
And yeah I'm going to make it work!!!
It's just the start have to make a lot of changes in the way
https://t.co/eFYkT2PZPW
Been away for a long time
January 2026 update:
-> Joined as a Python Developer intern at @sayheloai
-> Built & deployed multi-doc RAG system
-> Now working on Voice Translator MVP
Excited to keep building and learning!
GitHub: [https://t.co/oDLJiLlpzP]