Post

What actually happens when you call an LLM API? 🤔 I've been working on this visual explainer for a while! 🎬 Follow one request from client → gateway → router → GPU → streamed response. Plus KV caching, continuous batching, speculative decoding, and more. The entire journey in 5 minutes! 👇

sujee.dev

@sujee.dev

🎬 Watch in 4K: www.youtube.com/watch?v=vlfO... 🎤 Full #Hacktoberfest 2026 talk (45 min): www.twitch.tv/videos/29077... 📖 More details: sujee.dev/events/2026-...

youtube.comWhat Happens When We Call an LLM API? One request, traced to the GPU and backYouTube video by sujee_dev

27/10/26, 11:05 pm1 repost3 likes

sujee.dev

@sujee.dev

12h

🛠️ Behind the scenes: Part of my open-source Visual Stories series, using AI-assisted animations to explain complex technical concepts. Source code and production details on GitHub. Check it out, remix it, and share! 🚀 github.com/sujee/visual...

github.comvisual-stories/projects/tracing-llm-api-request at main · sujee/visual-storiesOpen-source, reproducible visual explainers for AI and software concepts. Watch them. Reproduce them. Remix them. - sujee/visual-stories