Identifying Token Costs Hiding in Your Agentic Loop
But cutting your runtime token burn is just the first problem.
Designing AI Agents That Can Self-Correct
With the vocabulary and the failure modes in place, here’s the build.
7 Chunking Strategies That Decide Whether Your RAG Works
Day 100 in production isn’t really about chunking strategies anymore.
Measuring Performance of Transformer Inference
This chapter is divided into eight parts; they are: • Metrics for LLM Inference • Measuring a Single Request • Warmup and Synchronization • Measuring GPU Work with CUDA Events • Measuring Memory Usage • Measuring Concurrent Requests • Multiple GPUs and Multiple Machines • Cost per Token The most common inference metrics are: • …
Static vs. Dynamic vs. Continuous Batching in LLM Inference
In this article, you will learn how static, dynamic, and continuous batching work in LLM inference, and why the differences between them matter at production…
Using a Transformer Model: From Training to Inference
This chapter is divided into four parts; they are: • Autoregressive Generation • Prefill and Decode • A Simple KV Cache • Memory Usage of the KV Cache A decoder-only transformer model predicts the next token from the tokens that came before it.
Decoding Strategies and Output Control
This chapter is divided into nine parts; they are: • Reading Logits from a Model • Greedy Decoding • Temperature Sampling • Top-$k$ Sampling • Nucleus Sampling • Repetition Penalties • Beam Search • Stop Conditions • Structured Output Constraints The model returns a vector of logits for every position in the input sequence.
The End-to-End Agentic AI Pipeline
In this article, you will learn the seven architectural components that separate a production-grade agentic AI system from a demo script, and how each one…
Ollama vs. LM Studio vs. llama.cpp: Which Local AI Runtime Should You Use in 2026?
In this article, you will learn how Ollama, LM Studio, and llama.cpp differ across the dimensions that matter most to practitioners, and how to choose…
From CUDA to MLX: How K-Search Brings Decades of Kernel Expertise to Apple Silicon
Figure 1: CUDA-to-MLX optimization translation map. CUDA optimization knowledge can be translated into architecture-native MLX strategies rather than copied instruction-for-instruction. We face a new epoch in computing. Hardware is changing rapidly — not just faster GPUs, but a growing range of chips from different vendors, each with its own architecture and often tailored to specific …
Read more “From CUDA to MLX: How K-Search Brings Decades of Kernel Expertise to Apple Silicon”