Topic

AI and agents

Engineering, operating and evaluating AI systems, coding agents and model infrastructure.

Blog posts
166
Collections
03

What are you trying to do?

Choose a reading sequence, from the initial decision to implementation and verification.

Coding agents and developer tools

Improve repository context, isolate work, manage skills and verify generated code.

  1. The Bottleneck Was Never Intelligence. It Was Context.
  2. Git Worktrees vs Jujutsu for AI Coding Agents: 2026 Decision Guide
  3. WikiSkill: How Agents Evolve SKILL.md from Experience
  4. QA for AI-Generated Code

Evaluation and production operations

Design the workflow, enforce authorization, budget evaluation and define rollout gates.

  1. AI Agent Design Patterns: Start Simple, Verify Actions
  2. Enterprise MCP Authorization Architecture
  3. When Is an LLM Eval Worth Building? Cost, ROI, and Trusting the Judge
  4. AI Agent Pilot in 30/60/90 Days

Local AI and deployment

Compare APIs with local inference, budget operations and assess hardware constraints.

  1. When Local Models Beat APIs: A Break-Even Calculator for EU Companies
  2. Self-Hosting LLMs in the EU: When Open Weights Actually Pay Off
  3. Colibri Runs GLM-5.2 on Consumer Hardware. Here Is the Catch.

Memory, RAG and company knowledge

Choose the retrieval architecture, estimate costs and verify permissions and freshness.

  1. RAG vs Fine-Tuning vs Long-Context 2026
  2. What an Internal AI Assistant Actually Costs in the DACH Region (2026)
  3. RAG Production-Readiness Checklist for EU Companies

Models and routing

Understand decision models, compare providers, check calibration, then plan a pilot.

  1. OpenAI Decisions API: Confidence, Refusals and Routing
  2. Cloudflare Clef vs Jev: Pricing, Benchmarks and Migration
  3. AnyJev: LLM Calibration and Option-Order Bias
  4. AI Agent Pilot in 30/60/90 Days

Explore this topic

80 posts

Agent engineering

Coding agents, MCP, context systems, evaluation and the controls required for dependable automation.

Collection →
8 posts

Decision models

Compare decision-model providers, handle typed answers, calibrate probabilities and ground bounded decisions in web evidence.

Collection →
78 posts

Models and infrastructure

Model selection, inference economics, local deployment, compression and serving architecture.

Collection →

Start here

Latest in this collection

Reviewed examples feed a compact classifier with automatic-decision and human-review paths AI & Agents

Small model fine-tuning: Airbnb’s production lesson

A specialist classifier needs more than a cheap inference call. Evaluate training data, accepted errors, fallback, full costs and the path to production.

Kevin Riedl
Read
A large euro symbol above segmented cost markers AI & Agents

Odoo AI credits: budget a reviewed quotation workflow

Budget the quotation queue before increasing volume. Separate Odoo credits, subscription uplift, initial review and correction work in a runnable synthetic ledger.

Kevin Riedl
Read
A shared starting point branches into 3 labeled paths AI & Agents

Odoo Helpdesk AI: classify, draft or escalate?

Separate ticket labels, internal reply drafts and human handoff. Use proposed support cases to test permissions and missed escalations.

Kevin Riedl
Read
A web evidence stream splits into bounded decisions in Wavect teal and violet AI & Agents

Parallel Decisions API: RAG for Decision Models in Practice

Turn current web evidence into bounded answers. A practical vendor-check example, with shared citations, optional confidence, batching costs and source limits.

Kevin Riedl
Read
Brackets framing an ontology derivation certificate in teal and violet AI & Agents

How to Validate Open Ontologies Derivation Certificates

Run a reproducible graph test, distinguish a forged-proof rejection from a checker failure, and inspect what a smaller RAG context still supports.

Kevin Riedl
Read
Separate paths for crawl requests and extracted records AI & Agents

Does Scrapling Save Scraped Items in Checkpoints?

A local pause, resume and replay test shows which records survive, when JSON exports replace earlier items, and which data a checkpoint saves.

Kevin Riedl
Read

Complete article directory

  1. Small model fine-tuning: Airbnb’s production lesson
  2. Odoo AI credits: budget a reviewed quotation workflow
  3. Odoo Helpdesk AI: classify, draft or escalate?
  4. Parallel Decisions API: RAG for Decision Models in Practice
  5. How to Validate Open Ontologies Derivation Certificates
  6. Does Scrapling Save Scraped Items in Checkpoints?
  7. TencentDB Agent Memory: Is SQLite Enough for Your AI Agent?
  8. Cue Turn Detection: Phone Numbers, Benchmarks and Limits
  9. RTK AI 1.2: Real Token Savings, Setup and Licensing
  10. Google SAM is now Agent Mesh: MCP tools across machines
  11. Anatomy for Claude Code: Interactive SVGs for Documentation
  12. EdgeQuake: PostgreSQL GraphRAG for MCP agent memory
  13. BitNet on CPU: 2B4T, real RAM needs and the 100B claim
  14. SuperPenguin: AI Cost per PR, Customer and Feature
  15. Claude Dashboards + Motion: From KPIs to Video
  16. OpenAI Intelligent UI vs OpenUI: Generative UI for Business Apps
  17. Wally by RunAnywhere: Cost per Accepted Coding Task
  18. Reducto for German Invoices: Measure Human Review Effort
  19. OpenAI Decisions API: Confidence, Refusals and Routing
  20. Claude Model Router: What Switches the Model, and When?
  21. Cloudflare Clef vs Jev: Pricing, Benchmarks and Migration
  22. Context Language Models vs Compaction: What to Pilot
  23. Caveman 3.0 for Claude Code: Local Input Compression, Recovery and Benchmarks
  24. LiteLLM Lens: Agent Trace Analysis with SQL and APIs
  25. SmythOS Studio Self-Hosting: Docker Setup, Costs and Limits
  26. DeerFlow 2.2: Docker Setup, Sandboxes and Memory
  27. CLM-8B Self-Hosting: vLLM, Action Cache and Verifiers
  28. AnyJev: LLM Calibration and Option-Order Bias
  29. LiteAgents SDK: Per-Turn Routing, Setup and Migration
  30. mcp-memory-service: Shared Memory for Claude Code and Cursor
  31. Claude Opus 5.5: Best Uses, Prompts and Effort Settings
  32. Laya and Jev for Business: 6 Practical Workflows
  33. Laya vs Jev: What the Benchmarks Mean for AI Startups
  34. Google AX: Suspend/Resume, Restart Costs and Self-Hosting
  35. WikiSkill: How Agents Evolve SKILL.md from Experience
  36. Tencent Octop Review: What Builders Actually Get for Free
  37. Wigolo Review: Local Web Intelligence for AI Agents
  38. Jev AI Review: Decision Models for Agent Workflows
  39. Supermemory: AI Agent Memory, RAG and Local Setup
  40. AI Agent Design Patterns: Start Simple, Verify Actions
  41. Claude Mods: Setup, Function Hooks and Security
  42. Voicebox: Local Voice Cloning, Dictation and MCP Setup
  43. Valyu’s 0.6B Multi-Agent Router: Results, Limits and When to Train One
  44. OpenAI Agents API Review: Migration, Costs and Data Controls
  45. Spotify shunt Review: Setup, Savings and Limits
  46. OpenBot Review: Self-Hosted AI Coworkers, Costs & Controls
  47. SwarmLLM Review 2026: Browser P2P LLM Inference Across Phones and Laptops
  48. Ramp Inspect Architecture 2026: Background Coding Agents at Scale
  49. Model Hardware Standard: Enterprise Guide to Physical AI
  50. Fonio AI Review 2026: Pricing, API, GDPR & Build vs Buy
  51. Mosaic (YC S26) Review: Shared Memory for Team AI Agents
  52. Ripwire Review 2026: AI Repo Context Without Embeddings?
  53. Atomic Multi-File Edits for AI Coding Agents: The Semaprax Lesson
  54. Phonely Alma Review: Is the Voice LLM Ready for Production?
  55. Utopia Review: Temporal Knowledge Graph for Enterprise
  56. AI Agent Knowledge Transfer: Use Frontier Models Once, Then Scale Cheaper
  57. claude-rotate: One Proxy for Multiple Claude Max Accounts
  58. NVIDIA PAIR Review: Local AI Routing and the AMD Gap
  59. Feynman Review: Is This Open-Source AI Research Agent Ready for Teams?
  60. Obscura Browser Review: Claims, Limits and Production Fit
  61. Claude Code Design System: 4 Parts for On-Brand UI
  62. Tencent Hy4 Preview Review: What Changes with the Full Hy4 Release?
  63. PageLM Review: Self-Hosting and Commercial Use
  64. ChatGPT Can Now Log In Without Seeing Your Password
  65. M6 Mac mini vs M5 Mac Studio for Local AI: Buyer’s Guide
  66. AI Agent Harness, Explained: The Reliability Layer Around an LLM
  67. FreeToken AI Review: Run Frontier MoE Models on Consumer GPUs?
  68. Darkbloom AI Review: Private Inference on Idle Macs
  69. Ox Alpha Revealed as GLM-5.3-Flash: What Changed
  70. Why Agent Edits Need Semantic Identity: Building SEMAPRAX in Rust
  71. LangChain Deep Agents Review: Is the Agent Harness Ready for Production?
  72. Pika Audio Models API Pricing: Is SFX Really 20x Cheaper?
  73. Thunder Compute’s $13M GPU Virtualization Bet: Enterprise Buyer’s Guide
  74. OpenViking Review: Filesystem Memory for AI Agents
  75. LLM-as-a-Verifier Explained: Architecture, Costs, and Production Fit
  76. AirLLM on 4 GB VRAM: How Layer-Wise Inference Really Works
  77. TrueForge Review: Is the Open-Source Agent Harness Production-Ready?
  78. Agent-Readable Websites: llms.txt, Markdown Mirrors and What Breaks
  79. Qwen3.8-27B: Self-Hosted Computer-Use Agents Without Exporting Screenshots
  80. Localized URL Slugs vs hreflang: Why We Keep One English Slug
  81. Can an AI Agent Use Your Product, or Only Read About It?
  82. Graft Review 2026: Do Agent Repo Maps Belong in Git?
  83. Codag for Coding Agents: Tool-Output Compression and Cost
  84. Smarter Token Usage: Measure Cost per Accepted Coding Task
  85. DeepSeek Harness Review: Is the Plugin Stack Production-Ready?
  86. Netflix’s vLLM and Triton Stack: 7 Production Lessons
  87. Alibaba OpenSandbox: Self-Hosting and Production Limits
  88. Transformers.js Browser AI: When Local Inference Belongs in Your Product
  89. Cloudflare Kitesurf Review: Cost, Limits and Production Fit
  90. GitHub Spec Kit Review: Is It Worth the Process?
  91. Internal AI Agent Marketplace: A 2026 Enterprise Build Guide
  92. How to Self-Host LiteLLM in Production: 2026 Guide
  93. AI-Ready Company Wiki: Architecture and Build Guide
  94. Is Linux the Best OS for AI Agents? A 2026 Infrastructure Guide
  95. Does Claude Watermark Text? The 2026 API Answer
  96. OpenKB Review: Knowledge Compiler vs RAG
  97. Unsloth Desktop Review: A Private Local AI Workstation?
  98. NeMo Switchyard 0.2: Agent Model Routing Without Training?
  99. Firecrawl AnyDoc Review: 14 Formats to Markdown
  100. MCP Cloud vs Manufact Cloud: MCP Hosting Guide
  101. Muse Glimmer 30B: Is Meta’s Local Agent Model Production-Ready?
  102. How to Make AI Writing Sound Human with Agent Skills
  103. NVIDIA NOOA Review: Are Object-Oriented Agents Production-Ready?
  104. OmniRoute AI Routing: Setup and Production Checklist
  105. Strix AI Pentesting: 30-Day Pilot and Buying Guide for 2026
  106. Hyperagent Review: Cloud AI Agents Without a Server
  107. Gemini Robotics 2: Whole-Body Control and the Pilot Decision
  108. Hark Handoff Review: The Agent That Actually Clicks
  109. Meta Muse Code Pricing: Is the Contributor Tier Safe for Client Code?
  110. pdf-inspector Review: Route PDFs Before OCR
  111. Agent Reach Review: Costs, Security and Real Limits
  112. Cloudflare Wallets for AI Agents: What Is Live?
  113. Local Multimodal AI Coding Assistant: Voice, OCR and Privacy
  114. DeepSeek V4 Flash 0731 on One AI PC: What Actually Works?
  115. YC QM Agent Review: Is Quartermaster Ready for Work?
  116. jcode vs Claude Code: Is the Rust Harness Worth Switching To?
  117. Lightpanda Browser for AI Agents: Production Guide
  118. Graph Engineering for AI Agents: When Does a Knowledge Graph Pay Off?
  119. Multi-Model AI Coding Agent Stack: A Team Buying Guide
  120. Is MCP Stateless Now? Your Server Migration Checklist
  121. MCP Is Not a Security Boundary: Protect Agent Data
  122. Fine-Tune Gemma 4 Free with Unsloth and Colab
  123. Can AI Agents Talk to Each Other? A Band Setup Guide
  124. Taalas HC1 Review: Is a Hardwired LLM ASIC Worth It?
  125. LeanCTX Technical Field Report: 64.1% Less Context
  126. llmfit Guide: Which Local LLM Fits Your Hardware?
  127. CLIProxyAPI: Run GPT-5.6 Sol Inside Claude Code
  128. Miso TTS Self-Hosted vs API: Cost, Latency and VRAM Reality
  129. Enterprise MCP Authorization Architecture
  130. Compress RAG Vectors to 2 Bits: Is Data-Oblivious Quantization Production-Ready?
  131. LMCache Reported 13.7x Lower Mean TTFT with a Shared KV Cache
  132. Lossless LLM Weight Compression vs 8-bit GGUF: What Is Ready for Production?
  133. Meterless Review 2026: Is This AI Agent Context Layer Ready?
  134. NVIDIA Nemotron 3.5 ASR: Is Free Self-Hosted STT Ready for Voice Agents?
  135. Claude Code vs OpenCode: Which Costs Less for a Team in 2026?
  136. Kimi K3 for EU Companies: API Cost, Data Risk, and a Pilot Plan
  137. Mesh LLM Review: Can One Large LLM Run Across Multiple Computers?
  138. Graphify Review 2026: Is a Codebase Knowledge Graph Worth It?
  139. Bonsai 27B Review: Can a 27B LLM Really Run on a Phone?
  140. Soofi S: Is Germany’s Sovereign LLM Ready for Business?
  141. AI Pilot Kill-or-Scale Scorecard: 12 Metrics to Check After 30 Days
  142. T3MP3ST Review 2026: Can It Replace a Penetration Test?
  143. Colibri Runs GLM-5.2 on Consumer Hardware. Here Is the Catch.
  144. Open Knowledge Format (OKF): The Enterprise Guide
  145. AI Agent Cost per Action: Why Agentic Workflows Blow Up Token Bills
  146. When Local Models Beat APIs: A Break-Even Calculator for EU Companies
  147. LLM Cost Calculator 2026: Cost per Task, Not Cost per Token
  148. Pxpipe Review: Can Images Cut Claude Code Token Costs 60%?
  149. Cheaper Per Token Can Still Cost More Per Task
  150. How to Use Claude Fable 5.2 in Claude Code
  151. Anthropic Says No Open-Weights Ban. The Cost War Is Still Real.
  152. What an Internal AI Assistant Actually Costs in the DACH Region (2026)
  153. LLM Gateways Compared 2026: LiteLLM vs OpenRouter vs Portkey vs RouteLLM
  154. Self-Hosting LLMs in the EU: When Open Weights Actually Pay Off
  155. Open-Weight LLMs 2026: DeepSeek, Qwen, Kimi, GLM, Llama and Muse
  156. The Bottleneck Was Never Intelligence. It Was Context.
  157. ChatGPT Enterprise vs Copilot vs Custom RAG
  158. MCP vs RAG vs Agent Skills vs Custom GPTs
  159. How to Cut LLM Token Costs in 2026
  160. AI Agent Pilot in 30/60/90 Days
  161. Permissions-First RAG over SharePoint, Confluence, Drive
  162. RAG Production-Readiness Checklist for EU Companies
  163. When Is an LLM Eval Worth Building? Cost, ROI, and Trusting the Judge
  164. Why AI Agent Projects Get Cancelled
  165. RAG vs Fine-Tuning vs Long-Context 2026
  166. LLM API Costs 2026. Architecture Shift
Inbox, without the noise

Get the next AI and agents field note

One concise email when we publish. No tracking pixels, and no inbox filler.