Featured
Fable Without Fable: Sakana Fugu Ultra's Big Bet
Sakana's most interesting move is selling learned orchestration as a frontier-model substitute, with Fable-class pressure and very different production risks.
Gen α AI · Field notes for AI builders
Depth over hype, for people who bet on AI.
Evidence-first analysis of agentic systems, model evaluation, and the economics of AI software. We read the system card, find the primary source, and tell you what actually changed — and what didn't.
Evidence over vibesDepth over volumeHonest about uncertainty
130Deep dives published
9Evergreen pillar guides
BiweeklyThe field briefing
Editor’s picksNew here? These are the pieces we’d hand you first.
01 · AI Tools
AI Coding Tools in 2026: The Power-User Field Guide
The gap between demo and production is the harness you build around the model, not the model you license.
Pillar guide → 02 · Search & GEOGenerative Engine Optimization: How to Earn AI Citations
Search is becoming synthesis. If ChatGPT, Perplexity, and Google's AI Overviews don't cite you, you're invisible, and…
Pillar guide → 03 · Agents & HarnessesAgent Harness Engineering and Agentic Loops: 2026 Field Guide
Execution loops, externalized state, and verification gates now matter more than raw model IQ. Here's how the agents…
Pillar guide → 04 · AI ToolsAI Coding Agent Economics: Real ROI and Cost per Pull Request
Frontier labs now ship more AI-written code than human-written code, but the viral ROI numbers are wrong. Here is the…
Pillar guide →The latestFresh analysis, published continuously — the full archive lives in the rail and the pillars below.
AI EconomicsAI Inference Hardware Has a New Cost Bottleneck
The Nvidia question is now a workload-matching problem: memory bandwidth, utilization, and latency SLOs decide the real inference bill.
10 minJune 23, 2026→
Model EvaluationLLM Evaluation Breaks When Teams Trust One Score
A production eval program needs offline gates, calibrated human judgment, and live monitoring tied to the failures that cost you money.
9 minJune 23, 2026→
AI FrontiersEU AI Act GPAI Transparency Code: Ship the Controls
A legal-looking Code becomes a release checklist once you map each Article 53 duty to artifacts, owners, and audit trails.
12 minJune 23, 2026→
AI FrontiersSmall Models Are Taking Over the Sovereign AI Stack
The practical path to AI sovereignty now runs through distillation, quantization, and deployable open-weight models instead of frontier-model procurement theater.
10 minJune 23, 2026→
AI FrontiersYour ML Team Probably Doesn't Need a Feature Store
Feature stores are assumed in modern MLOps, but the real cutoff is production complexity, not ambition.
12 minJune 23, 2026→
AI FrontiersRunning Local AI Models Just Crossed the Line
The 2026 shift is less about one miracle model and more about open weights, quantization, unified memory, and inference runtimes finally landing at the same time.
12 minJune 23, 2026→
Memory & ContextLong Context vs RAG: Stop Chunking at the Right Time
Million-token windows changed the default, but retrieval still wins when citations, query volume, and latency matter.
11 minJune 22, 2026→
Model EvaluationAI Coding CLI Telemetry Has an SSD Problem
A Codex SQLite logging bug turns telemetry from an abstract privacy concern into a measurable workstation endurance risk.
10 minJune 22, 2026→
AI FrontiersConductor LLMs Make Model Choice a Product Lever
The winning AI product architecture is shifting from picking one frontier model to owning the policy that routes work across many.
12 minJune 22, 2026→
AI FrontiersFrontier Model Access Can Vanish. EU Teams Need a Plan
The Anthropic Fable/Mythos shutdown turned model choice into a continuity problem for EU engineering teams.
11 minJune 22, 2026→
Memory & ContextVector Database Comparison: Speed Is the Trap
Production RAG teams should choose a vector store by operating model, filter shape, and migration triggers, not by a vendor latency chart.
12 minJune 22, 2026→
AI EconomicsAI Video Generator Comparison: Pick What Ships
The practical video stack decision is no longer model quality alone; it is usable seconds, editing drag, rights clearance, and where the clip has to ship.
13 minJune 22, 2026→
Explore the pillarsNine durable guides that organize everything we publish.
AI Tools15 pieces
AI Coding Tools in 2026: The Power-User Field Guide
The gap between demo and production is the harness you build around the model, not the…
Explore →Search & GEO9 pieces
Generative Engine Optimization: How to Earn AI Citations
Search is becoming synthesis. If ChatGPT, Perplexity, and Google's AI Overviews don't cite…
Explore →Agents & Harnesses17 pieces
Agent Harness Engineering and Agentic Loops: 2026 Field Guide
Execution loops, externalized state, and verification gates now matter more than raw model…
Explore →AI Economics13 pieces
AI Coding Agent Economics: Real ROI and Cost per Pull Request
Frontier labs now ship more AI-written code than human-written code, but the viral ROI…
Explore →Model Evaluation17 pieces
Evaluating AI Models and Agents: The 2026 Field Guide
Why static leaderboards lost authority, and how to build an eval program that survives…
Explore →Memory & Context14 pieces
Context Engineering for AI Agents: Memory, RAG & MCP
Why the context window, not the prompt, is the real bottleneck, and how to engineer…
Explore →Security & Safety8 pieces
Securing AI Agents and LLM Apps: The 2026 Threat Model
Why indirect prompt injection, tool-mediated exfiltration, and rogue agents now define LLM…
Explore →Models & Releases11 pieces
AI Models 2026: The Mid-Year Frontier and Open-Weight Map
How the open-weight cluster closed the gap, why reasoning became the default, and which of…
Explore →AI Frontiers26 pieces
AI Frontiers 2026: Diffusion Models, Multimodal AI & More
A practitioner's map of frontier AI in mid-2026, where independent measurement finally…
Explore →