How a GPU Actually Works
The intuition an LLM engineer needs. Understand techniques like quantization, speculative decoding, and continuous batching in one place.
Lire
A Cheaper Model Does Not Imply a Cheaper Turn
The practical implications of model routing, clearly explained.
Lire
How Production LLMs Reason Better At Inference Time
8 techniques, explained visually.
Lire
Continuous Batching in LLMs
The technique behind vLLM's 23x throughput jump and the default scheduler in every serving engine.
Lire
![[Hands-on] Audio RAG with 200x Cheaper Vector DB Costs](https://substackcdn.com/image/fetch/$s_!Qscc!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F51c48d12-3f06-4d05-a92a-cb2851e3b5b6_1096x1078.gif)
[Hands-on] Audio RAG with 200x Cheaper Vector DB Costs
...while also outperforming OpenAI and Cohere.
Lire
Karpathy's Full Agentic Engineering Lifecycle using Google's Agents-CLI
...explained as step-by-step guide.
Lire
How to Query Billion+ Rows on Postgres Without the Overhead
...explained as a full setup guide.
Lire
A 10-week Roadmap to Run LLMs in Production
...covered with hands-on resources.
Lire
![[Hands-on] Build Semantic Search Inside Your Database Without an Embedding Pipeline](https://substackcdn.com/image/fetch/$s_!D_AD!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F1843328d-aac8-46cd-8d4d-a0aed939af4b_3110x1992.png)
The Missing Piece of Agent Self-Improvement
...explained step-by-step with code.
Lire
![[Hands-on] How to Serve 5 Models On One GPU](https://substackcdn.com/image/fetch/$s_!szXp!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F939f26ce-5485-4313-9783-e6ab7f94db08_1200x415.png)
[Hands-on] How to Serve 5 Models On One GPU
How small specialized models are changing inference infrastructure, and why serving them efficiently takes more than standard serving frameworks.
Lire
Why Your Agent Remembers Everything and Understands Nothing
Building a pattern recognition layer for memory in production.
Lire
The Hands-on AI Engineer Playbook to Build RAG Apps for Production
Why RAG latency is a prefill problem, not a retrieval problem.
Lire
Build a Stock Market Research Agentic Workflow
...using a no-code drag-and-drop builder.
Lire
6 Automatic Optimization Methods for LLM Systems
...explained visually and with practical tradeoffs.
Lire