Daniel Keller — TechProduct & Technology leadership, GenAI development, and side projects.https://danielkeller.com/en-usWiring Gemma 4 Into Claude Code — A Practical Local-Plus-Cloud Setuphttps://danielkeller.com/tech/gemma4-in-claude-code/https://danielkeller.com/tech/gemma4-in-claude-code/A step-by-step guide to running Gemma 4 locally via LM Studio and calling it from inside a Claude Code session — for offline reasoning, batch processing, privacy-sensitive data, and zero-cost first-pass drafts that hand off to Claude when you need the heavy hitter.Wed, 08 Apr 2026 00:00:00 GMTTechclaude-codegemma4lm-studioollamalocal-llmhybrid-aiopenai-compatibleapple-silicontutorialClaude Code Ultraplan — Planning Moves to the Cloudhttps://danielkeller.com/tech/claude-code-ultraplan/https://danielkeller.com/tech/claude-code-ultraplan/Anthropic just shipped a hidden Claude Code feature called Ultraplan. Instead of planning locally, your session is offloaded to a multi-agent Opus 4.6 swarm in the cloud that builds a richer, faster, more structured plan — and then teleports it back to your terminal for execution.Tue, 07 Apr 2026 00:00:00 GMTTechclaude-codeultraplanplanningopus-4-6multi-agentcloudagenticMy Favourite Claude MCPs for macOShttps://danielkeller.com/tech/favourite-claude-mcps-macos/https://danielkeller.com/tech/favourite-claude-mcps-macos/A living list of the Model Context Protocol servers I actually use every day on my Mac — what they do, why they earned a spot in my setup, and how to install each one in under five minutes.Sun, 05 Apr 2026 00:00:00 GMTTechclaudemcpmacosclaude-codeclaude-desktopxmindobsidianproductivityFrom Terminal to Factory — Running Headless AI Agents 24/7https://danielkeller.com/tech/headless-agents/https://danielkeller.com/tech/headless-agents/The step-by-step guide to transitioning from local Claude Code sessions to persistent, headless backend agents you control from your phone. Dispatch, remote control, agent teams, and event-driven automation.Fri, 03 Apr 2026 00:00:00 GMTTechclaude-codeheadlessdispatchremote-controlagent-teamsautomationdevopsRunning Gemma 4 Locally: Frontier-Class, Zero Costhttps://danielkeller.com/tech/gemma4-local-coding/https://danielkeller.com/tech/gemma4-local-coding/Google's Gemma 4 26B runs locally on Apple Silicon, scoring within single digits of GPT-5.2 and Claude Opus 4.5 on reasoning benchmarks — reshaping the local-plus-cloud AI coding workflow.Thu, 02 Apr 2026 00:00:00 GMTTechgemma4local-llmapple-siliconollamacoding-assistantopen-weightsapache-2hybrid-aibenchmarksThe Perfect CLAUDE.md for Enterprise Software Engineeringhttps://danielkeller.com/tech/perfect-claude-md/https://danielkeller.com/tech/perfect-claude-md/How I use CLAUDE.md files to encode engineering excellence standards — turning tribal knowledge into enforceable, AI-augmented guardrails for building enterprise-grade SaaS.Wed, 25 Mar 2026 00:00:00 GMTTechclaude-mdclaude-codeengineering-standardsenterprisesaasagentic-developmentclean-codetestingci-cdRunning Local LLMs — Qwen3-Coder, DeepSeek, MLX, and Apple Siliconhttps://danielkeller.com/tech/local-llm-setup/https://danielkeller.com/tech/local-llm-setup/A practical guide to running powerful language models locally on Apple Silicon — model selection, MLX vs Ollama, quantization trade-offs, and when local beats cloud.Sun, 15 Mar 2026 00:00:00 GMTTechlocal-llmqwen3deepseekmlxollamaapple-siliconprivacydeveloper-toolsAgentic Development Patterns — Building Software with AI Agentshttps://danielkeller.com/tech/agentic-development-patterns/https://danielkeller.com/tech/agentic-development-patterns/Practical patterns and workflows for agentic software development using Claude Code, Cursor, and local LLMs — from parallel workstreams to overnight agent factories.Sun, 01 Mar 2026 00:00:00 GMTTechagentic-developmentclaude-codecursorllmdeveloper-toolsworkflowNutriSpan — Gamifying Longevity Sciencehttps://danielkeller.com/tech/nutrispan-app/https://danielkeller.com/tech/nutrispan-app/A React Native app that turns longevity optimization into a game — with a Longevity Impact Score, 1,000+ food database, daily checklists, streaks, and achievements grounded in peer-reviewed science.Sun, 01 Mar 2026 00:00:00 GMTTechreact-nativeexpohealthlongevitygamificationzustandtypescriptSortFlex — 3D Configurator for Custom Holders & Mountshttps://danielkeller.com/tech/sortflex/https://danielkeller.com/tech/sortflex/A browser-based 3D design tool for custom holders, organizers, and mounts. Drag and drop everyday items onto a base plate, then 3D print at home or order a printed version shipped to your door.Sun, 01 Mar 2026 00:00:00 GMTTechreact-three-fibercadquery3d-printingcloudflarevitereactAI-Driven Development — Agentic Coding in 2026https://danielkeller.com/tech/ai-driven-development/https://danielkeller.com/tech/ai-driven-development/How agentic AI has transformed software development — from pair programming to autonomous coding agents, overnight agent factories, and vibe coding.Sun, 15 Feb 2026 00:00:00 GMTTechaisoftware-developmentclaude-codecursoragentic-developmentvibe-codingAI Product and Tech Resourceshttps://danielkeller.com/tech/ai-resources/https://danielkeller.com/tech/ai-resources/A curated collection of the most important AI tools, frameworks, and resources for product and tech leaders — updated for the agentic era.Sun, 01 Feb 2026 00:00:00 GMTTechairesourcestoolsclaude-codecursoragentic-developmentlocal-llmThree-Tier AI Architecture — Deterministic Rules, ML Models, and LLM Agentshttps://danielkeller.com/tech/three-tier-ai-architecture/https://danielkeller.com/tech/three-tier-ai-architecture/A practical framework for deciding when to use hard-coded rules, trained ML models, or LLM agents in production AI systems — with real examples from enterprise SaaS.Sun, 01 Feb 2026 00:00:00 GMTTechai-architecturellmmachine-learningproductionsystem-designPrompting in the Agentic Era — From Prompt Engineering to System Designhttps://danielkeller.com/tech/ai-prompting/https://danielkeller.com/tech/ai-prompting/How prompting has evolved from clever tricks to system-level specification — CLAUDE.md files, structured outputs, tool-use patterns, and when prompting still matters.Tue, 20 Jan 2026 00:00:00 GMTTechaipromptingllmclaude-mdstructured-outputtool-useagentic-aiAI in a Nutshell — Models, Concepts, and the 2026 Landscapehttps://danielkeller.com/tech/ai-in-a-nutshell/https://danielkeller.com/tech/ai-in-a-nutshell/A concise overview of AI models, fundamental concepts, and the current state of AI engineering — from frontier models to local LLMs and agentic systems.Thu, 15 Jan 2026 00:00:00 GMTTechaimachine-learningllmfoundation-modelsai-engineeringagentic-aimcpThe Overnight Agent Factory — Autonomous Development with Claude Code & Cursorhttps://danielkeller.com/tech/overnight-agent-factory/https://danielkeller.com/tech/overnight-agent-factory/How I run parallel AI agents overnight to build production software while I sleep.Thu, 15 Jan 2026 00:00:00 GMTTechclaude-codecursoragentic-developmentminimaxApplied AI & ML Ops — From Traditional MLOps to Agentic Infrastructurehttps://danielkeller.com/tech/applied-ai-mlops/https://danielkeller.com/tech/applied-ai-mlops/How to bridge the gap between AI research and production — from classical MLOps to MCP servers, tool-use APIs, RAG pipelines, and agentic system infrastructure.Sat, 10 Jan 2026 00:00:00 GMTTechaimlopsdataopsmachine-learningai-engineeringmcpragproduction