Skip to content
View voyager2005's full-sized avatar
🥰
Falling in love with coding (Part 2)
🥰
Falling in love with coding (Part 2)

Block or report voyager2005

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 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
voyager2005/README.md

Akshat G

Machine Learning Engineer • Passionate Researcher • Artificial Intelligence • Data-Centric AI

"Building intelligent systems that understand data before learning from it."


akshat@tech:~$ whoami
Akshat G

akshat@tech:~$ pwd
/home/akshat
Experience
├── Shell India Markets Private Limited
├── Sensera
└── Manipal Institute of Technology

akshat@tech:~$ tree research
research
├── HAViT
│   ├── Hierarchical Multi-label Classification
│   └── Galaxy Zoo Benchmark
├── BTCV-DDPM
│   ├── Diffusion Pretraining
│   └── Medical CT Representation Learning
├── SemiSupervised-Spine
├── MoNuSeg
├── UNet-Benchmarks
└── Polyp-Segmentation

akshat@tech:~$ echo $CURRENT_FOCUS
• Foundation Models
• Data-Centric AI
• Representation Learning
• Production AI Systems
• Computer Vision
akshat@tech:~$ cat philosophy.txt

When I first started learning AI, I thought better models came from adding more layers,
more parameters, or more compute.
The more I learnt, the more I found myself asking a different question.
Instead of making models bigger, can we help them understand the world better?

That curiosity has slowly pulled me towards representation learning,
data-centric AI, and understanding what models actually perceive
rather than simply optimizing benchmark numbers.

I don't have all the answers yet, but I hope I never stop asking the questions.

Long term, I hope to contribute towards scalable,
interpretable and data-centric AI systems capable of solving
real-world problems.

Technical Toolkit

Languages

Python Java C++ JavaScript HTML CSS Bash

AI / Machine Learning

TensorFlow PyTorch OpenCV

Development

Git Docker Linux VS Code

Cloud & Deployment

AWS Azure GCP

akshat@tech:~$ apt install technical-skills
Reading package lists... Done
Building dependency tree... Done
Resolving dependencies... Done

Installing:
✔ Python                   ✔ TensorFlow
✔ OpenCV                   ✔ Docker
✔ Linux                    ✔ CUDA
✔ SQL                      ✔ Git
✔ Computer Vision          ✔ Deep Learning
✔ Representation Learning  ✔ MLOps

Done.

akshat@tech:~$ ls programming/
Python/      Java/
C++/         SQL/
TensorFlow/  PyTorch/
CUDA/        Docker/
Git/         Linux/

akshat@tech:~$ status
Research Papers ............... ███████=====
Projects ...................... ████████====
Engineering ................... ██████████==
Coffee ........................ ████████████
GPU Memory .................... Never Enough
Sleep ......................... 0% (see: Coffee)

akshat@tech:~$ history
1  Discover AI . . . . . . . . 2018
2  Learn Java  . . . . . . . . 2019
3  Build first project . . . . 2020
4  Join research . . . . . . . 2024
5  Publish papers  . . . . . . 2024
6  Build production systems  . 2025
7  Keep learning . . . . . . . Present

akshat@tech:~$ exit
logout

Thanks for stopping by.

Pinned Loading

  1. cpp-neural-net cpp-neural-net Public

    My from scratch implementation of CNN on c++ without using any ML libraries... Its for things like this that i love computer science and this concept of AI, the feeling of churning out a NN at 3am …

    C++

  2. an-atom an-atom Public

    I love physics, so why not simulate an atom! This repo is a direct inspiration for one of the youtube videos i saw earlier, and this is my implementation of simulating an atom down to the core of q…

    Python

  3. coe-autonomous-mobility coe-autonomous-mobility Public

    Python 1