---
title: "Benzahra Karima (@kaghima) | Hashnode"
description: "Computer Science Engineer | Full-Stack AI & DevOps. Read the latest articles by Benzahra Karima on Hashnode."
url: "https://hashnode.com/@kaghima"
---

@kaghima

# Benzahra Karima

@kaghima · Joined June 2025

Computer Science Engineer | Full-Stack AI & DevOps

[Website](https://kalima-ocr.com/)

### [From Notebook Jail to Production: Scaling LLMs on NVIDIA Blackwell](https://kaghima.hashnode.dev/from-notebook-jail-to-production-scaling-llms-on-nvidia-blackwell?utm_source=hashnode&utm_medium=feed)

Tired of Jupyter Notebook jail? I deployed Qwen 3.5 on NVIDIA Blackwell. Here is how I built a production-ready, scalable LLM infrastructure stack.

3 min read · [kaghima.hashnode.dev](https://kaghima.hashnode.dev/from-notebook-jail-to-production-scaling-llms-on-nvidia-blackwell?utm_source=hashnode&utm_medium=feed)

[Mar 5](https://hashnode.com/posts/from-notebook-jail-to-production-scaling-llms-on-nvidia-blackwell/69a8d12d3b896f7ad68e488a)

### [Step-by-Step — Compiling vLLM from Source on NVIDIA Blackwell (GB10)](https://kaghima.hashnode.dev/step-by-step-compiling-vllm-from-source-on-nvidia-blackwell-gb10?utm_source=hashnode&utm_medium=feed)

Tired of Docker crashes on your DGX Spark? Learn how to natively compile vLLM for Blackwell (GB10) using CUDA 13. A guide for high-performance AI infr

3 min read · [kaghima.hashnode.dev](https://kaghima.hashnode.dev/step-by-step-compiling-vllm-from-source-on-nvidia-blackwell-gb10?utm_source=hashnode&utm_medium=feed)

[Mar 4](https://hashnode.com/posts/step-by-step-compiling-vllm-from-source-on-nvidia-blackwell-gb10/69a7871656428acc6f226d5a)
