NARENDRA KUMAR VADAPALLI

AI LEADER • VISUAL DESIGN ENTHUSIAST

A software engineer with a balanced left and right brain, specializing in pipeline development, DevOps, graphics tools, and site reliability.

Read My Blog View Portfolio Resume

Featured Projects

View All Projects →

Credits in Media

Following are the entertainment / media projects I contributed technically

Details →

Graduation work

Projects from my Masters program in Computer Graphics and Game Technology at University of Pennsylvania

Details →

Latest Blog Posts

Read All Posts →

Aug 9, 2026

Part 6: Why Deep Networks Die: Weight Initialization (He/Xavier), LayerNorm, and Residual Connections

Why do 100-layer networks crash with NaNs? Learn how Xavier/Kaiming variance scaling, LayerNorm/RMSNorm, and ResNet identity shortcuts keep deep architectures alive.

Read Post →

Aug 8, 2026

Part 5: Inside the Learning Engine: Forward Pass, Backpropagation, and Dynamic Autograd

Peer inside PyTorch's core learning engine: how predictions flow forward, loss errors walk backward via the chain rule, and autograd DAGs record every operation.

Read Post →

Aug 7, 2026

Part 4: Demystifying Activation Functions: Why Neural Networks Need Non-Linearity, Types, and Real-World Use Cases

Why are linear models useless for complex data? Explore non-linear space warping, dead ReLU activations, GELU, and SwiGLU.

Read Post →