Stephen Diehl

Posts tagged
ai

MLIR Part 14 - GPT-2 on the GPU
We Live in the Dependently Typed Future Now
No, Transformers Won't End the Human Race lol
MLIR Part 13 - Running GPT-2
MLIR Part 12 - Fusion, Bufferization and Execution
MLIR Part 11 - Matrix Multiplication and Attention
MLIR Part 10 - Compiling Transformer Kernels
MLIR Part 9 - A Python Frontend for MLIR
Can Opus 4.6 do Category Theory in Lean?
MLIR Part 8 - GPU Compilation with MLIR
MLIR Part 7 - Transformers
MLIR Part 6 - Specializing Python with E-graphs
MLIR Part 5 - Neural Networks
MLIR Part 4 - Linear Algebra in MLIR
MLIR Part 3 - Affine Dialect and OpenMP
The Stochastic Code Monkey Theorem
It Would Be Good if the AI Bubble Burst
MLIR Part 2 - Memory in MLIR
Fine-tuning With Tool Calling
Program Synthesis: The λ in the Machine
What is AI market actually worth?
Interfacing MCP with Combinatorial, Convex, and SMT Solvers
Remote MCP Servers
Adventures in Symbolic Algebra with Model Context Protocol
Using CUDA Deep Neural Network (cuDNN) in Python
AI Slopocalypse 2027
Attention Wasn't All We Needed
MLIR Part 1 - Introduction to MLIR
MLIR Part 0 - Installing MLIR
SGLang on AMD MI300X
The FlashAttention CUDA Kernel Line by Line
Training with GRPOTrainer
Process Reward Models
Introduction to Fine-tuning Large Language Models
Setting up a Nvidia GH200 for Development
Hymba : Mamba × Transformers
Tiny GraphRAG (Part 2)
Tiny GraphRAG (Part 1)
The Limitations of RAG
AGI Discourse as Kayfabe
Monte Carlo Tree Search
Fine-tuning with ORPO and Unsloth
A Rapid Tutorial on Unsloth
Word Calculators are Useful, but they're not Smart
Going Deep on DeepSpeed
GPT-2 in One Function
View all tags