# Build for Agents

> A newsletter keeping track of the monumental ideas and methodologies behind building the next generation of software that can be used natively by agentic systems. Only key milestones, no filler.

Build for Agents is a curated newsletter that explores the fundamental shifts in software development as we move towards agent-native systems. The site documents key research papers, white papers, protocols, and methodologies that are shaping how we build software for AI agents and large language models.

The newsletter is authored by [Mark Narvidas](https://www.narvidas.com) and focuses on practical insights for developers, product teams, and organisations looking to understand and implement agent-native software architectures.

## Core Content

- [Main Newsletter Page](https://buildforagents.com/): The primary newsletter content with chronological entries covering key developments in agent-native software
- [RSS Feed](https://buildforagents.com/feed.xml): Machine-readable feed for newsletter subscribers and content aggregation

## Key Resources

- [Andrej Karpathy: Software Is Changing (Again)](https://www.youtube.com/watch?v=LCEmiRjPEtQ): Video presentation on Software 3.0 and natural language as a programming interface
- [A practical guide to agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf): OpenAI's comprehensive guide for building agents
- [Agents Companion Whitepaper (PDF)](https://www.kaggle.com/whitepaper-agent-companion): Google's developer guide for advanced agent topics
- [Model Context Protocol](https://www.anthropic.com/news/model-context-protocol): Anthropic's open standard for secure AI tool connections
- [The /llms.txt file](https://llmstxt.org/): Jeremy Howard's proposal for standardising LLM-friendly website documentation
- [Attention Is All You Need](https://arxiv.org/abs/1706.03762): The landmark 2017 transformer architecture paper
- [Bret Victor's "DBX" Talk](https://worrydream.com/dbx/): 2013 talk on future programming paradigms and goal-oriented systems