ai
82 notes tagged ai in the AI agents topic. How agents are built, tuned and run: loops, harnesses, MCP, skills and coding agents, with what broke and what we measured.
Writing
- 18 Oct 20263,016 delegations to agents: the long briefs cost the mostEssay
- 18 Oct 2026Harness engineering notes from a team running agents all dayEssay
- 8 Oct 2026Five layers to tune before you fork your agentAnalysis
- 7 Sep 2026What the Engelberg report got right about usAnalysis
Show all 22
- 3 Sep 2026Claude Code 2.1.259 turned our Read() deny rules into a Bash tripwireAnalysis
- 3 Sep 2026The debug loop: make the agent measure the browser before it touches the CSSAnalysis
- 18 Aug 2026DeepSeek harness, dissectedAnalysis
- 14 Aug 2026What stays valuable when anyone can build anythingEssay
- 27 Aug 2025E2b breakdownAnalysis
- 30 Jul 2025Cline breakdownAnalysis
- 28 Jul 2025Better engineeringEssay
- 9 Jun 2025Engineers in the AI landscape: architects, tinkerers, planners, and the vibersEssay
- 2 Jun 2025Composing the future, part 1: agent basics, LangGraph and AutogenAnalysis
- 2 Jun 2025Composing the future, part 2: OpenAI's agent tools, MCP and fast-agentAnalysis
- 2 Jun 2025Composing the future, part 3: Mastra and CrewAIAnalysis
- 2 Jun 2025Composing the future, part 4: LlamaIndex and the framework face-offAnalysis
- 2 Jun 2025Composing the future, part 5: agent patterns, open problems and how to chooseAnalysis
- 29 May 2025SkillmaxingEssay
- 18 Oct 2026We stopped testing engineers on speedEssay
- 13 Aug 2026The tellsEssay
- 21 Jun 2025Speed beats perfectionism in AI adaptationEssay
- 27 May 2025Vetting proven talentsEssay
Reports
- 24 Aug 2026Making our site AI-agent-ready, 50 to 93Field report
- 19 Aug 2026Our publish step took 110 seconds, then 78. Our release tab took three tries.Field report
- 10 Jul 2025Composing the Dwarves video pipelineField report
- 21 Nov 2024Building chatbot agent to streamline project managementField report
Show all 7
- 21 Nov 2024Building data pipeline for OGIF transcriberField report
- 14 Nov 2024Project reports system: a case studyField report
- 18 Oct 2026We called ten technology arcs in June 2025. eight of the 60 experiments we planned got built.Field report
Research notes
- 18 Apr 2025How to use Claude to build the system promptResearch notes
- 10 Jul 2024How we crafted the OGIF summarizer bot to streamline weekly knowledge-sharingResearch notes
- 29 May 2023How we created an AI powered interview system using OpenAI's ChatGPTResearch notes
- 3 Sep 2025How knowledge work organizes itselfResearch notes
Show all 51
- 3 Sep 2025The five stages of learningResearch notes
- 8 Jul 2025Should AI agents transact without human confirmation?Research notes
- 23 Jun 2025AI agent explained: LLMs, workflows and agentsResearch notes
- 23 Jun 2025Prompt formulaResearch notes
- 19 Jun 2025Seeding your content for AI: a writer’s guideResearch notes
- 18 Jun 2025Stand up GEO tracking tools: a simple guideResearch notes
- 12 Jun 2025Generative engine optimizationResearch notes
- 2 Jun 2025Mapping the path to AGIResearch notes
- 27 Mar 2025Authorization for remote MCP servers over SSEResearch notes
- 27 Mar 2025Tool-level security for remote MCP servers, part 1: designing role-based access controlResearch notes
- 27 Mar 2025Tool-level security for remote MCP servers, part 2: enforcement, identity and practiceResearch notes
- 27 Jan 2025Let AI ask youResearch notes
- 27 Jan 2025Your AI approachResearch notes
- 19 Dec 2024Synthetic users in UX researchResearch notes
- 19 Dec 2024Working with legacy codeResearch notes
- 13 Nov 2024Natural language to database queries: Text-to-MongoDBResearch notes
- 15 Oct 2024Model selectionResearch notes
- 23 Sep 2024Prevent prompt injectionResearch notes
- 12 Jul 2024Streamlining internal tool development with managed LLMOps: A Dify case studyResearch notes
- 11 Jul 2024Building agent supervisors to generate insightsResearch notes
- 13 Jun 2024A grand unified theory of the AI hype cycleResearch notes
- 18 May 2023Choosing a vector database for LLM applicationsResearch notes
- 18 Oct 2026#5 Building is cheap now, getting heard is the hard partResearch notes
- 18 Jul 2025LSTMResearch notes
- 12 Jun 2025Staying visible in AI search with GEOResearch notes
- 12 Jun 2025Understanding Tidewave.aiResearch notes
- 25 Apr 2025Multi-MCP data integration - receiving data from other MCP serversResearch notes
- 2 Apr 2025Build custom AI agent with ElizaOSResearch notes
- 21 Nov 2024Quantization for large language modelsResearch notes
- 18 Oct 2024ReAct(reason + act) in LLMResearch notes
- 18 Oct 2024ReWOO: Reasoning without observation - a deeper lookResearch notes
- 4 Oct 2024Use cases for LLM applicationsResearch notes
- 1 Oct 2024The rise of AI applications with LLMResearch notes
- 11 Sep 2024Journey of thought prompting: Harnessing AI to craft better promptsResearch notes
- 8 Aug 2024What is Generative UI?Research notes
- 18 Jul 2024Function calling in AI agentsResearch notes
- 3 Jul 2024Proximal policy optimizationResearch notes
- 25 Jun 2024Explaining gradient descent in machine learning with a simple analogyResearch notes
- 2 May 2024Developing rapidly with generative AIResearch notes
- 9 Aug 2023A story map for building with LLMsResearch notes
- 10 Jul 2023Adversarial prompting in prompt engineeringResearch notes
- 29 Jun 2023LLM's accuracy - self refinementResearch notes
- 23 Jun 2023Reward modelResearch notes
- 22 Jun 2023Q learningResearch notes
- 9 Jun 2023Building blocks of a RAG pipelineResearch notes
- 5 Jun 2023Introduction to reinforcement learning and its application with LLMsResearch notes
- 18 May 2023Foundation models: the latest advancement in AIResearch notes
Case studies
- 7 Jul 2025AI-powered consulting sprints for rapid clarityCase studies
- 31 Mar 2025AI for fast and fair talent searchCase studies