CS & AI Engineer — Far Western University, Nepal
3rd semester CSIT student building things that matter — from neural networks to full-stack apps. Passionate about AI/ML, systems programming, and creating software with real impact.
I'm Milan Yadav, a 21-year-old Computer Science & IT student at Far Western University, Nepal. I build everything from terminal animations to full transformer architectures from scratch — because I believe the best way to understand something is to build it.
My academic focus spans OOP with C++, Operating Systems, Computer Architecture, and Discrete Mathematics. But outside class, I'm deep in PyTorch, training neural networks, building MiniGPT transformers, and experimenting with DCGANs and reinforcement learning agents.
I'm driven by a simple idea: software should solve real problems for real people. Whether it's a passive income SaaS tool or a model that understands language — I want to build it.
A character-level GPT transformer built from scratch using PyTorch. Implements multi-head self-attention, positional encoding, and autoregressive text generation.
Deep Convolutional GAN that generates synthetic images using adversarial training. Includes custom discriminator and generator architectures with training visualizations.
Deep Q-Network agent that learns to play environments using experience replay, target networks, and epsilon-greedy exploration strategy.
Fully connected neural network implemented using only NumPy — backpropagation, gradient descent, activation functions, and loss computation, no ML frameworks.
Full-stack grading management app built in React + C++ backend. Features grade tracking, GPA calculation, exception handling, and report generation.
Visual simulator for CPU scheduling algorithms — FCFS, SJF, Round Robin, and Priority. Includes Gantt charts, wait time and turnaround time calculations.
Whether you have a project in mind, a collaboration idea, or just want to connect — I'm always open to meaningful conversations. Response usually within 24 hours.