Writing on LLM

• The Problem is Prompt Debt,2026
• How Kimi K2 RL'ed Qualitative Data to Write Better,2025
• How Kimi K2 Became One of the Best Tool-Using Models,2025
• How Rewriting Training Data Improved Kimi K2's Performance,2025
• Cat Facts Cause Context Confusion,2025
• How to Fix Your Context,2025
• Prompts vs. Context,2025
• How Long Contexts Fail,2025
• The Drawbridges Go Up,2025
• Let the Model Write the Prompt,2025
• The Wisdom of Artificial Crowds,2025
• What We Mean When We Say “Think”,2025
• MCPs are APIs for LLMs,2025
• Building an Easier to Use FFmpeg With LLMs,2025
• A Gentle Intro to Running a Local LLM,2025
• Mistral Small & Human-Centric Benchmarks,2025
• Your Eval is More Important Than the Model,2025
• On Test-Time Compute: The New Game in Town,2024
• On Synthetic Data: How It's Improving & Shaping LLMs,2024
• Pipelines & Prompt Optimization with DSPy,2024
• Why LLM Advancements Have Slowed: The Low-Hanging Fruit Has Been Eaten,2024
• Transitioning from the Attention Era to the Automation Era,2024
• AI Search Apps Succeed Because of Their UI, Not their LLMs,2024
• Generating Descriptive Weather Reports with LLMs,2024
• All writing · All topics