ai tips
💸 Cost saving
Using ai models with APIs can be expensive and cost you > 300€/month, if you are not careful.
-
Switch models while coding.
-
Use smaller prompts.
-
Cancel subscriptions.
-
Consolidate to on LLM router service.
🦮 Guiding the LLM
Guiding the Coding Assistant towards satisfactory outcomes is therefore of ever-growing importance in our daily work.
3 critical measures that are required to work successfully in an AI-assisted coding setup:
- Well-structured Requirements: defines the destination.
- Tool-based Guard Rails: ensure we take the most direct route towards the destination.
- File-based Keyframing: create near-empty stub files, and the LLM can take over.
📜 References
- Senior Developer Skills in the AI Age: Leveraging Experience for Better Results
- Explore, expand, exploit
- LLM Agents are simply Graph — Tutorial For Dummies
- 2019-12-19 - Building Effective AI Agents \ Anthropic
- 2025-03-03 - From Design doc to code: the Groundhog AI coding assistant (and new Cursor vibecoding meta)
- 2025-02-16 - My LLM codegen workflow atm - Harper Reed’s Blog
- 2025-05-08 - Basic Claude Code - Harper Reed’s Blog
- 2025-05-15 - Amp is now available. Here’s how I use it.
- 2025-05-21 - Practical AI techniques for daily engineering work
- 2025-06-07 - Field Notes From Shipping Real Code With Claude
- 2025-06-08 - How I program with Agents
- 2025-06-12 - Don’t Build Multi-Agents
- 2025-06-20 - how i bring the best out of claude code - part 2
- 2025-06-23 - Context Engineering for Agents
- 2025-06-25 - How I keep up with AI progress (and why you must too)
- 2025-06-26 - How to Fix Your Context
- 2025-06-30 - Developing with GitHub Copilot Agent Mode and MCP
- 2025-07-13 - Context Engineering: Bringing Engineering Discipline to Prompts
- 2025-07-14 - Context Rot: How Increasing Input Tokens Impacts LLM Performance
- 2025-07-21 - The 3P Architecture: A Deep Dive into Software Agent Design (with Manus AI)
- Six principles for Production AI agents
- 2025-07-28 - AI for Coding: Why Most Developers Get It Wrong (2025 Guide)
- 2025-07-29 - Mastering agentic workflows with ADK: the recap
- 2025-08-08 - AI must RTFM: Why technical writers are becoming context curators
- 2025-08-14 - A persona-based approach to AI-assisted software development
- 2025-08-23 - Double your Productivity at Work using these 3 GenAI Prompt Strategies
- 2025-09-04 - Vibe Coding Our Way to Disaster
- 2025-09-15 - The quality of AI-assisted software depends on unit of work management
- 2025-09-30 - Which Table Format Do LLMs Understand Best?
- 2025-10-14 - Just Talk To It - the no-bs Way of Agentic Engineering
- 2025-10-15 - Understanding Spec-Driven-Development: Kiro, spec-kit, and Tessl
- 2025-10-30 - AI’s real superpower: consuming, not creating
- 2025-11-02 - What if you don’t need MCP at all?
- 2025-12-20 - You Don’t Need to Spend $100/mo on Claude Code: Your Guide to Local Coding Models
- 2026-01-04 - The gap between a Helpful Assistant and a Senior Engineer
- 2026-01-10 - Code Is Cheap Now. Software Isn’t
- 2026-01-11 - The biggest obstacle for engineer productivity in 2026
- 2026-01-13 - AI model choices 2026-01 - Kaushik Gopal’s Website
- 2026-01-13 - How to write a good spec for AI agents
- 2026-01-14 - Give your agent a laboratory
- 2026-01-13 - Running Claude Code dangerously (safely)
- 2026-01-27 - 10 Months of Agentic Coding: No Vibes or Slop Allowed
- 2026-02-02 - Coding agents & use cases
- 2026-02-05 - Context Engineering for Coding Agents
- 2026-02-12 - The AI hater’s guide to code with LLMs (The Overview)
- 2026-02-19 - Simple Made Inevitable: The Economics of Language Choice in the LLM Era
- 2026-02-28 - The 2-Minute Claude Code Upgrade You’re Probably Missing: LSP - Karan Bansal
- 2026-03-09 - Claude Code, Claude Cowork and Codex #5
- 2026-03-21 - Optimizing Skills
- My Agent Skill for Test-Driven Development - SaturnCI - Continuous Integration for Ruby on Rails
- 2026-04-07 - Vibe coding is not the same as coding assisted by AI
- 2026-08-15 - Deciphering Glyph :: The Futzing Fraction
Workflows
- Here’s how I use LLMs to help me write code
- 2025-09-24 - humanlayer/advanced-context-engineering-for-coding-agents
- 2025-10-09 - Superpowers: How I’m using coding agents in October 2025
- 2026-02-10 - How I Use Claude Code
- 2026-05-17 - How I use AI for software engineering — Chris Etheridge’s Blog
- 2026-08-11 - Chad Arimura | My agent setup
📏 Convention
- Writing Cursor Rules with a Cursor Rule
- Claude Code Best Practices - Anthropic
- How I use Cursor (+ my best tips)
- I use Cursor daily - here’s how I avoid the garbage parts
- You are using Cursor AI incorrectly…
- Literate Development: AI-Enhanced Software Engineering
- Senior Developer Skills in the AI Age: Leveraging Experience for Better Results
- AI Blindspots
- 2026-02-12 - 65 lines of Markdown
🤖 Agentic
- Design Patterns for Agentic AI and (Multi) AL Agents
- Claude Code Best Practices
- GPT-4.1 Prompting Guide
- Use some framework to build the agent , e.g. Pocket Flow.
- EnzeD/vibe-coding
- VIBE CODING - The Ultimate Guide with Resources
- LLMs are mirrors of operator skill
- 2025-06-12 - Agentic Coding Recommendations
- 2025-06-18 - Parallelizing AI Coding Agents
- 2025-07-29 - Mastering agentic workflows with ADK: the recap
- skillmatic-ai/awesome-agent-skills: The definitive resource for Agent Skills - modular capabilities revolutionizing AI agent architecture
- Awesome Agentic Patterns
- The Agent Skills Directory
Frameworks
- BMAD-METHOD: Breakthrough Method for Agile Ai Driven Development
- The AI framework that adds the engineering to prompt engineering
🔬 Code exploration
- A Missing Mental Model: Why AI doesn’t Understand Codebases
- Promoted his tool Giga AI to help LLM with less hallucination by creating project maps.
- How Windsurf Breaks Through the Celling for Retrieval
- Rerankers and Two-Stage Retrieval