Shawn Azdam

Agentic Coding · Jun 8, 2026 · 15 min read

Build Around the Agent, Not the API

The next AI engineering stack should treat the coding agent as the execution layer and the repository, tools, tests, permissions, and observability around it as the real product surface.

AI Workflows · Apr 4, 2026 · 11 min read

Reasoning Is the New Build Artifact

The cheapest AI workflow is not always the one that uses cheaper models everywhere. It is the one that turns expensive reasoning into a durable artifact, then lets lower-cost models execute against it.

Machine Learning · Jan 10, 2023 · 3 min read

Get Up and Running with NVIDIA GPUs on Ubuntu

I have been setting up multiple Python machine learning environments based on NVIDIA GeForce GPUs. Starting with a fresh Ubuntu installation, you will need to go through the installation of multiple NVIDIA packages to get your machine ready to communicate with the GPU(s) properly. The steps here are an overview of the installation process.

Machine Learning · Aug 29, 2021 · 3 min read

Install TensorFlow and PyTorch Together in a Single Environment

I have been recently dealing with projects requiring both the popular automatic differentiation frameworks together. Installing both TensorFlow and PyTorch in a single environment might be a challenge of its own kind.

Selected papers

CORE: Uncertainty-Gated Policy Optimization for Reliable Performance Under Distribution Shift

ReleaseMind: Autopoietic Risk-Constrained Release Note Generation

ManeuverGPT: Agentic Control for Safe Autonomous Stunt Maneuvers

Conditional-Flux Disk-Type Hysteresis Motor Without Axial-Flux Disk Holder

Notes from the event horizon

Build Around the Agent, Not the API

Reasoning Is the New Build Artifact

Get Up and Running with NVIDIA GPUs on Ubuntu

Install TensorFlow and PyTorch Together in a Single Environment