llms.txt Content
# Avinash Sangle — Claude Code & AI Expert from Pune
> AI Automation Engineer and Claude Code expert based in Pune, India. Building AI-powered developer tools, MCP servers, and DevOps automation. Specializes in Claude Code workflows, Model Context Protocol, and agentic engineering.
## Author
- Website: https://avinashsangle.com
- LinkedIn: https://www.linkedin.com/in/avinashsangle
- Twitter/X: https://x.com/avi_sangle
- GitHub: https://github.com/avisangle
- Location: Pune, Maharashtra, India
- Expertise: Claude Code, AI Automation, Model Context Protocol, DevOps, Generative AI
## Blog Posts
- [Getting Started with the ant CLI: Deploy Claude Agents](https://avinashsangle.com/blog/ant-cli-getting-started): Install the ant CLI, create your first managed agent, and deploy it in under 10 minutes. YAML version control, scripting patterns, and CI/CD.
- [Claude Code Cost Tracking: Monitor and Cut Your Spending](https://avinashsangle.com/blog/claude-code-cost-tracking): Track Claude Code costs with built-in commands, JSONL logs, ccusage, and 7 tips to cut spending by 50%.
- [Claude Managed Agents vs Agent SDK: Which Should You Use?](https://avinashsangle.com/blog/claude-managed-agents): Decision framework for Anthropic's new Claude Managed Agents beta vs the Agent SDK. Covers pricing, MCP integration, and when to pick each.
- [Claude Mythos: What It Means for Developers Who Can't Use It](https://avinashsangle.com/blog/claude-mythos-preview): Analysis of Anthropic's restricted Claude Mythos Preview release, Project Glasswing, benchmarks vs Opus 4.6, and practical guidance for developers.
- [Gemma 4 Models: Which One Should You Actually Use?](https://avinashsangle.com/blog/gemma-4-models-guide): Tested all 4 Gemma 4 model sizes locally. Practical guide to picking E2B, E4B, 26B MoE, or 31B Dense based on hardware and use case.
- [How I Write CLAUDE.md Files That Actually Work (2026)](https://avinashsangle.com/blog/claude-md-guide): Real production CLAUDE.md examples from