What Frontier Reasoning Models Change About Product Work
Posted 9 August 2025
What Context Engineering Really Means for Agents
12 min read
Understanding Chain-of-Thought Prompting in 2026
11 min read
A Working Guide to Few-Shot Prompting
13 min read
Zero-Shot Prompting, Explained Properly
10 min read
Evaluating Agents Without Guesswork
14 min read
What Makes a Model Agentic? A Technical Guide
15 min read
Automatic Reasoning and Tool Use, Step by Step
12 min read
Tree-of-Thought Prompting in Practice
13 min read
Decomposed Prompting: When to Reach for It
12 min read
The Five Levels of Agent Autonomy
14 min read
Raising Accuracy with Generated Knowledge
13 min read
ReAct Prompting: What Still Holds Up
13 min read
Recursive Prompting for Very Long Documents
10 min read
Least-to-Most Prompting, Demystified
11 min read
Active Prompting: Choosing What to Label
10 min read
Iterative Prompting Without the Drift
10 min read
Thornbury is generally available. We're giving away $1.4M in evaluation credits.
15 min read
The Reasoning Benchmarks That Actually Bite
15 min read
Self-Consistency Prompting, Measured
15 min read
Few-Shot or Fine-Tune? A Cost Read
8 min read
Zero-Shot Prompting at Production Scale
7 min read
What Eval Leads Should Know in 2026
16 min read
Scaling Reinforcement Learning Inside LLMs
10 min read
Test-Time Scaling, Without the Hype
10 min read