What Tobi Lütke believes
Distilled positions on specific topics, each backed by citations to their public posts. Older versions of a belief are preserved when the position evolves.
There is a strong focus on developing open-source AI orchestration frameworks and platforms for building context-engineered, production-ready LLM applications and agent workflows, with explicit control over retrieval, routing, memory, and generation.
- tobi starred deepset-ai/haystack: Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems.· github
- tobi starred langgenius/dify: Production-ready platform for agentic workflow development.· github
- tobi starred paperclipai/paperclip: Open-source orchestration for zero-human companies· github
- tobi starred BoundaryML/baml: The AI framework that adds the engineering to prompt engineering (Python/TS/Ruby/Java/C#/Rust/Go compatible)· github
AI is being leveraged to search, scrape, and interact with the web, providing an API for AI systems to gather and process information from online sources.
AI is seen as the fastest path to achieving full-stack observability, even for lean teams, suggesting its critical role in monitoring and understanding complex systems.
Entrepreneurship is the most resilient job in an era of rapid technological change, acting as a strategic hedge against automation and market shifts.
Local, personal AI systems, like a 'Second Brain' built from extensive personal data, can significantly boost productivity by augmenting LLM interactions and continuously improving, especially in high information overload environments.
There is significant interest in and development of open-source alternatives for AI tools, including ChatGPT alternatives, AI orchestration frameworks, and local AI engines that can run various models on diverse hardware.
- tobi starred garrytan/gbrain: Garry's Opinionated OpenClaw/Hermes Agent Brain· github
- tobi starred rcarmo/piclaw: I'm going to build my own OpenClaw, with blackjack... and bun!· github
- tobi starred janhq/jan: Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.· github
- tobi starred deepset-ai/haystack: Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems.· github
- tobi starred infiniflow/ragflow: RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs· github
- tobi starred mudler/LocalAI: LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.· github
Companies scaling to massive transaction volumes inevitably face challenges like exhausting 32-bit and then 64-bit ID namespaces, underscoring the need for proactive planning in distributed systems to avoid retroactive regrets over minor storage savings.