https://blog.krizaka.com/ 2026-10-16T12:00:00Z The Krizaka blog: how we build sovereign AI in Montréal, what it really costs, and what our products do on real documents. Sober, factual, in English and French. https://blog.krizaka.com/posts/anatomy-of-an-ai-prompt/ 2026-10-16T12:00:00Z 2026-10-16T12:00:00Z Oussama Abid

A six-word prompt, then three layers of detail, on my Mac with the same seed. What gets better, what gets worse, what a negative prompt can do, and what I didn't prove.

I had the story in my head before I made the pictures. You type six words, the model gives you a three legged player, and you learn you should have written a better prompt. It's the story you read everywhere. I only wanted to illustrate it. The model didn't play along. What it showed me was less dramatic, and more useful. So here is what I actually got, with the prompt, the seed and the time under every picture. Every picture was made on my MacBook Pro (M1 Max, 64 GB) with the image engine's command line. One picture per prompt, a single seed (2026). This is a demonstration, not a benchmark: I https://blog.krizaka.com/posts/local-first-ai-platform-spring-ai-2-and-the-mac/ 2026-10-16T12:00:00Z 2026-10-16T12:00:00Z Oussama Abid

A Mac out of breath, a ClassCastException that killed the chat, models that became database rows. What I learned running Orazaka on a laptop.

On 12 October my Mac hit the wall. The system load swung between 100 and 260, and tools started failing with ENFILE : the system refused to open a single file more. Several agent sessions were working in parallel, each with its own native builds, Docker containers or mobile simulator. All of it on a MacBook Pro M1 Max with 64 GB of memory. That Mac is the first server of Orazaka, the AI engine I'm building. The idea behind Orazaka is simple: run AI on a machine you control. For me, that machine is my laptop first. Here's what it taught me. An AI platform is a lot of cooks in a small kitchen Ev https://blog.krizaka.com/posts/what-image-and-video-generation-costs/ 2026-10-16T12:00:00Z 2026-10-16T12:00:00Z Oussama Abid

A Mac, a hosted API and a rented H100, on the same pictures and clips. The seconds, the dollars, the mistakes that cost money, and a small calculation for when renting a GPU pays off.

During our last video batch, the final clip came off the GPU and the pod kept running. For about two hours, an H100 rented at US$3.99 an hour sat waiting for work that would never come. It had a watchdog meant to shut it down after a while without jobs. The watchdog didn't, because destroying the pod meant calling RunPod's API, and the pod had no key to do that. A guard loop running on my Mac is what finally stopped it. Those two hours cost between $7 and $8, on a GPU bill of $23.23. That stings, and it's a good place to start talking about what generating images and video really costs. Every https://blog.krizaka.com/posts/why-krizaka-builds-sovereign-ai-in-montreal/ 2026-10-16T12:00:00Z 2026-10-16T12:00:00Z Oussama Abid

A GPU rented in Montréal, a Quebec privacy law, and three plain questions to ask any AI tool. What sovereign means at Krizaka, what it costs, and what is still unsolved.

This week our demo videos were generated on an NVIDIA H100 rented from RunPod. In the rental log, the pod's region is called CA MTL 1. Montréal. The machine that made our pictures was running in my own city. That was not the plan. I wanted a 48 GB GPU at DigitalOcean, in Toronto, where our databases already live. There was never one in stock in that region, the only one our private network reaches. So we took the H100 in Montréal. Looking at that region name, it struck me as a good place to start explaining what we do. First, who is writing My name is Oussama. I'm a Java and Spring engineer, a