# dev>notes > Programming tutorials, AI engineering, developer tools, and honest model comparisons. Site: https://mohitkhare.me Author: Mohit Khare Feed: https://mohitkhare.me/feed.xml Sitemap: https://mohitkhare.me/sitemap.xml ## About dev>notes is a technical blog covering AI engineering, coding tools, prompt engineering, and developer productivity. Written by Mohit Khare, a developer based in Europe with 6+ years of experience. Content focuses on real-world testing and opinionated takes rather than surface-level overviews. ## Key Topics - AI model comparisons and benchmarks - AI coding assistants (Claude Code, Cursor, Copilot, Devin, Windsurf, OpenClaw) - AI configuration files (CLAUDE.md, SOUL.md, AGENTS.md, .cursorrules) - Prompt engineering and prompt libraries - Agentic coding and autonomous development - MCP servers and AI tool integrations - RAG, vector databases, and LLM infrastructure - Obsidian workflows for developers - Developer productivity and workflows - Programming best practices ## Interactive Tools - [AI Cost Calculator](https://mohitkhare.me/tools/): Enter token usage, compare monthly costs across GPT-5.4, Claude, Gemini, Grok, DeepSeek, Qwen. - [AI Model Picker Quiz](https://mohitkhare.me/tools/): 5 questions to find the right AI model for your workflow. - [AI Token Counter](https://mohitkhare.me/tools/token-counter/): Paste text, see token counts and costs per model. Runs client-side. - [CLAUDE.md Generator](https://mohitkhare.me/tools/claude-md-generator/): Generate production-ready CLAUDE.md configs for Claude Code projects. - [AI Model Timeline](https://mohitkhare.me/ai-timeline/): 52 model releases from Nov 2022 through Mar 2026, filterable by company, country, and license. ## Important Articles - [Claude Opus 4.7: Better at Coding, Worse at Trust](https://mohitkhare.me/blog/claude-opus-4-7-first-look/): Opus 4.7 review with full benchmark table, tokenizer cost analysis, adaptive thinking issues, community backlash, and Qwen 3.6 coverage. - [Claude Mythos: Everything We Know About Anthropic's Most Powerful Model](https://mohitkhare.me/blog/claude-mythos-everything-we-know/): Anthropic's April 7 2026 model. 93.9% SWE-bench, found a 27-year-old OpenBSD zero-day, escaped its own sandbox during testing. Restricted to Project Glasswing partners. $25/$125 per million tokens. Full breakdown with primary source citations. - [Best AI Models April 2026: Claude Mythos Just Broke the Tier List](https://mohitkhare.me/blog/ai-models-compared-april-2026/): April 2026 ranking of every major AI model. Mythos sits in a new M tier above S. Opus 4.6 and GPT-5.4 Pro share S tier. Qwen3.5 promoted to A. - [How to Actually Read LLM Benchmarks Without Getting Fooled](https://mohitkhare.me/blog/how-to-read-llm-benchmarks/): What MMLU, HumanEval, and GPQA actually measure, why providers cherry-pick scores, and what to look at when choosing a model. - [Claude Code Hooks: The Feature Nobody's Talking About](https://mohitkhare.me/blog/claude-code-hooks-guide/): Auto-format, lint, test, and block dangerous commands with Claude Code hooks. Full settings.json examples. - [The Definitive Guide to AI Developer Tools in 2026](https://mohitkhare.me/blog/ai-developer-tools-definitive-guide/): Every AI developer tool ranked - IDEs, CLI tools, models, hosting, and deployment with honest recommendations. - [Obsidian + Claude Code: The Developer Second Brain I Actually Use](https://mohitkhare.me/blog/obsidian-claude-code-developer-setup-2026/): Full setup for using Obsidian as a dev knowledge base with Claude Code integration. - [OpenClaw vs Claude Code: I Tested Both for a Month](https://mohitkhare.me/blog/openclaw-vs-claude-code-2026/): 331K GitHub stars vs Anthropic's official CLI - which AI coding tool actually ships code. - [SOUL.md vs CLAUDE.md vs AGENTS.md: AI Config Files Compared](https://mohitkhare.me/blog/soul-md-vs-claude-md-vs-agents-md-2026/): Comparison of AI configuration file formats and which one matters most in your project. - [Best AI Models March 2026: Every Model Ranked](https://mohitkhare.me/blog/ai-models-compared-march-2026/): GPT-5.4, Claude Opus 4.6, Gemini 3.1 Pro, Grok 4.20, Qwen3.5 ranked from daily use. - [Vibe Coding: Real Talk About AI-Assisted Development](https://mohitkhare.me/blog/vibe-coding-real-talk/): Honest take on the vibe coding trend and when it works vs when it falls apart. - [Building with AI Agents in 2026](https://mohitkhare.me/blog/building-with-ai-agents-2026/): Practical guide to building production applications using AI agents. - [AI Experiments 2026: What I Learned](https://mohitkhare.me/blog/ai-experiments-2026/): Results from running multiple AI experiments across coding, writing, and automation. - [My Exact Claude Code Configuration](https://mohitkhare.me/blog/my-claude-code-config/): The exact config, prompts, and workflows used with Claude Code daily. - [Experiment: Can AI Build a Full App Autonomously?](https://mohitkhare.me/blog/ai-experiment-autonomous-coding/): Testing whether an AI agent can build a complete application from a spec without human intervention. - [Claude Opus 4 Deep Review: Best Model for Code?](https://mohitkhare.me/blog/claude-opus-4-review/): Thorough review of Claude Opus 4 for real-world coding tasks. - [Best AI Prompts for Developers in 2026](https://mohitkhare.me/blog/best-prompts-2026/): The prompts that actually make a difference in daily developer workflows. - [How I Became a 10x Engineer with AI](https://mohitkhare.me/blog/10x-engineer-with-ai/): How AI tools are redefining what it means to be a highly productive engineer. - [AI-First Development: Why I Stopped Writing Boilerplate](https://mohitkhare.me/blog/ai-first-development/): How shifting to an AI-first approach eliminates repetitive coding work. - [Lessons from Running AI Agents in Production](https://mohitkhare.me/blog/ai-agents-production-lessons/): Real lessons learned from deploying and maintaining AI agents in production systems. - [Windsurf vs Cursor in 2025: Which AI IDE Wins?](https://mohitkhare.me/blog/windsurf-vs-cursor-2025/): Side-by-side comparison of the two leading AI-native code editors. - [My Prompt Library: 30 Prompts I Use Weekly](https://mohitkhare.me/blog/prompt-library-developer/): A curated collection of battle-tested prompts for development tasks. - [Using AI to Refactor Large Codebases](https://mohitkhare.me/blog/ai-refactoring-large-codebases/): Techniques for using AI to safely refactor large, complex codebases. - [Devin vs Claude Code: AI Coding Agents Compared](https://mohitkhare.me/blog/devin-vs-claude-code/): Head-to-head comparison of autonomous coding agents for real development work. - [MCP Servers: Connecting AI to Everything](https://mohitkhare.me/blog/mcp-servers-guide/): Guide to Model Context Protocol servers and how they extend AI capabilities. - [Claude Opus 4 vs GPT-5: The Big Comparison](https://mohitkhare.me/blog/claude-vs-gpt5-comparison/): Detailed comparison of the two top-tier AI models for coding and reasoning. - [Building AI Agents: A Practical Guide](https://mohitkhare.me/blog/building-ai-agents/): Step-by-step guide to building functional AI agents from scratch. - [Agentic Coding Explained: Why It's Different](https://mohitkhare.me/blog/agentic-coding-explained/): What agentic coding means and how it changes the developer workflow. - [My Claude Code Setup: Terminal AI That Actually Works](https://mohitkhare.me/blog/claude-code-setup/): Complete guide to setting up and configuring Claude Code for terminal-based AI development. - [Building RAG from Scratch (No LangChain)](https://mohitkhare.me/blog/rag-from-scratch/): How to build a retrieval-augmented generation system without framework dependencies. - [Running Local LLMs with Ollama: My Guide](https://mohitkhare.me/blog/local-llms-ollama-guide/): Setup guide for running open-source language models locally with Ollama. - [The Best System Prompts I've Written](https://mohitkhare.me/blog/best-system-prompts/): Collection of effective system prompts for different AI use cases. - [Prompt Engineering Guide](https://mohitkhare.me/blog/prompt-engineering-guide/): Practical prompt engineering techniques that produce better AI outputs. - [Vector Databases Explained Simply](https://mohitkhare.me/blog/vector-databases-explained-simply/): Plain-language explanation of vector databases and when to use them. - [LLM Hallucinations: Real Examples and How to Handle Them](https://mohitkhare.me/blog/llm-hallucinations-real-examples/): Documented examples of LLM hallucinations with strategies to detect and prevent them.