KrausCrypto

A normal person and a crackhead on the street are coalescing into a point of singularity. The schizophrenic guy flailing around has so much in common with the normal person TikTok dancing in the street.

Wow I really 100x my Productivity. AMA?

Equinox is the only place in the world where you can spend rent money and still not be able to get a machine

@designcmmnty @Praha37v Hey man are you looking for a new job at a startup?

Isometric Designs 🎞️ By @Praha37v

@businessbarista Claude

I want to start a community dedicated to Claude Code. It’s become the gateway drug to coding and experiencing the power of AI for tons of people. This will be a space for people to share killer use cases, agentic workflows, proven prompts, and connect with other CC obsessives. Comment “Claude” if you want to join.

@omarsar0 @krauscrypto/agents-to-maximize-human-efficacy-193e0928de27" target="_blank" rel="nofollow noopener">medium.com/@krauscrypto/a… I wrote about it first here back in february. I call it the action space

Anthropic just posted another banger guide. This one is on building more efficient agents to handle more tools and efficient token usage. This is a must-read for AI devs! (bookmark it) It helps with three major issues in AI agent tool calling: token costs, latency, and tool composition. How? It combines code executions with MCP, where it turns MCP servers into code APIs rather than direct tool calls. Here is all you need to know: 1. Token Efficiency Problem: Loading all MCP tool definitions upfront and passing intermediate results through the context window creates massive token overhead, sometimes 150,000+ tokens for complex multi-tool workflows. 2. Code-as-API Approach: Instead of direct tool calls, present MCP servers as code APIs (e.g., TypeScript modules) that agents can import and call programmatically, reducing the example workflow from 150k to 2k tokens (98.7% savings). 3. Progressive Tool Discovery: Use filesystem exploration or search_tools functions to load only the tool definitions needed for the current task, rather than loading everything upfront into context. This solves so many context rot and token overload problems. 4. In-Environment Data Processing: Filter, transform, and aggregate data within the code execution environment before passing results to the model. E.g., filter 10,000 spreadsheet rows down to 5 relevant ones. 5. Better Control Flow: Implement loops, conditionals, and error handling with native code constructs rather than chaining individual tool calls through the agent, reducing latency and token consumption. 6. Privacy: Sensitive data can flow through workflows without entering the model's context; only explicitly logged/returned values are visible, with optional automatic PII tokenization. 7. State Persistence: Agents can save intermediate results to files and resume work later, enabling long-running tasks and incremental progress tracking. 8. Reusable Skills: Agents can save working code as reusable functions (with SKILL .MD documentation), building a library of higher-level capabilities over time. This approach is complex and it's not perfect, but it should enhance the efficiency and accuracy of your AI agents across the board. anthropic. com/engineering/code-execution-with-mcp

@benln I’m in

We're booking out a cafe on October 21st in Soho Grab coffee, Cursor credits, co-work, and meet the team Let me know if you'd like to come by

It’s all about being as close to a builder as possible now in tech. People managers telling other engineers what to do is so antiquated. If you’re a lead and not writing code what are you doing???

AI Code generation is way more effective at smaller startups. It’s great for writing net new code or rewriting entire features. At larger companies when you’re taking features from 80% to 100% accuracy. The engineer with lots of experience becomes super valuable.

Most companies now are barely scratching the surface of what’s possible with LLMs. Once you start to develop the mindset around how to bring determinism into a largely stochastic process it opens up the mind to what’s possible

@blakeandersonw How all companies calculate ARR these days

GPT-5 might be the worst update Open AI has ever released. How is it possible a supposedly better model can't even adhere to simple return only this schema responses? Page Numbers omitted by GPT-5 but maintained perfectly using GPT-4.1. This needs to be fixed. @sama

Coming to studios near you soon!

@kearneyy Poland!

wibes

Agents without memory of me and what I care about and all the context around me are just not as useful We are so early it is not yet table stakes but it will be

@garrytan Have you heard of useloma.com

@wenbinf Yes, for once the small startups are the ones that can reap the benefits.

great news for small software businesses!