ModelCraft
Subscribe
Sign in
Home
Podcast
Notes
Chat
Archive
About
The $27k Weekend: Why Your Agent OOMs at 62% GPU Utilization
Agents Aren’t Just Expensive API Calls. They’re stateful workflows on very pricey hardware. Old problem in new clothes but when the clothes cost you a bomb.
READ THE LATEST
Latest
Top
Discussions
The Real Reason LLMs Feel Slow
The Mental Model Every AI Engineer Needs: Prefill + Decode Latency Breakdown
Feb 12
•
Abi Aryan
21
2
2
Agentic AI's death is real BUT the Survivors are building Empires
Thanks to everyone who was able to join my lightning talks on Maven.
Feb 10
•
Abi Aryan
6
1
The Hidden GPU Crisis in AI Infrastructure
For everyone who couldn't join my ODSC West 2025 talk, I promised you details over a blogpost. Well, here we go - Most AI companies waste HALF their…
Nov 4, 2025
•
Abi Aryan
16
1
From Tensors to Teraflops: A Mental Model for GPU Engineering for LLMs
I promised to share some material on GPU Engineering after my last talk on Fundamentals of GPU Orchestration at Luma.
Sep 27, 2025
•
Abi Aryan
28
4
Why the hell do I need to learn GPU Engineering for AI Systems?
I promised lessons for the 4 roles in my last Maven session. Here's the first one for those interested in going beyond AI Engineering.
Aug 5, 2025
•
Abi Aryan
22
1
AI Careers in 2025+
Thank you for everyone who was able to join my Maven Talk!
Jul 30, 2025
•
Abi Aryan
6
1
How GPUs became relevant?
Tracing the Breakthroughs That Shrunk Computers and Supercharged AI
Apr 12, 2025
•
Abi Aryan
4
2
1
See all
ModelCraft
Making learning about ML tools and services fun for everyone. Hate spam. Just 1 letter a month.
Subscribe
ModelCraft
Subscribe
About
Archive
Recommendations
Sitemap
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts