Uge 40

How Elon Musk solves problems: First principles thinking explained | Lex Fridman Podcast Clips

A Lex Fridman podcast clip discussing Elon Musk's approach to solving problems from first principles.

What is First Principles Thinking?

An explanation of first-principles thinking, with examples of separating assumptions from fundamental facts.

AI PM OS for Claude Code & Codex

A product-management toolkit of agent skills, frameworks and guided workflows for AI coding assistants.

The First Principles Method Explained by Elon Musk

A Kevin Rose interview segment in which Elon Musk explains first-principles thinking.

extend-hq/jevbox

A self-hosted document library that organises uploads and answers questions using hierarchical search and citations to source material.

Introducing System One Models & Jev - TypeSafe AI Blog

TypeSafe introduces Jev and its approach to models that produce structured decisions for software automation.

Quick start - TypeSafe AI

A walkthrough for making structured requests to TypeSafe models through its playground and API.

Models - TypeSafe AI

Reference information for TypeSafe's Jev models, their API identifiers and operating limits.

Introduction - TypeSafe AI

An introduction to TypeSafe models that evaluate supplied context and return typed answers for applications.

State - TypeSafe AI

An explanation of how to structure the context supplied to TypeSafe models as text or JSON.

console.typesafe.ai/

langchain-ai/langgraph

A framework for coordinating AI agents through workflows with state, branching and persistent execution.

e2b-dev/E2B

An execution environment that gives AI agents access to tools in isolated sandboxes.

agno-agi/agno

Tools for creating, operating and managing platforms built around AI agents.

langfuse/langfuse

An observability and evaluation platform for inspecting AI application traces and assessing output quality.

confident-ai/deepeval

A testing framework for evaluating the behaviour and quality of applications powered by language models.

browser-use/browser-use

Software that lets AI agents interact with websites through a browser.

mastra-ai/mastra

A TypeScript framework for developing applications that use AI agents and related workflows.

topoteretes/cognee

A memory platform that helps AI agents retain and retrieve information across interactions.

pydantic/pydantic-ai

Python tools for building typed AI applications, including agents, voice interactions, image generation and embeddings.

getzep/graphiti

A toolkit for constructing knowledge graphs that update as AI agents encounter new information.

garrytan/gstack

Garry Tan’s collection of specialised coding-agent workflows covering planning, design, review, quality assurance and delivery.

GorvGoyl/Clone-Wars

A catalogue of open-source recreations of familiar websites, with links to code, demonstrations and technology choices.

oraios/serena

An MCP toolkit for finding and editing code using its semantic structure.

aquasecurity/trivy

A security scanner for vulnerabilities, exposed secrets and configuration problems across containers, infrastructure and repositories.

langwatch/langwatch

A platform for tracing and evaluating AI agents, with controls for routing and governance.

gitleaks/gitleaks

A utility for detecting credentials and other secrets accidentally included in code or version history.

SWE-bench/SWE-bench

A benchmark that evaluates whether AI systems can resolve software issues drawn from real GitHub projects.

yamadashy/repomix

Packages a repository into a single document so an AI tool can inspect its code and structure together.

gitbutlerapp/gitbutler

A desktop client for working with Git version control, implemented with Tauri, Rust and Svelte.

bridgecrewio/checkov

Checks infrastructure definitions, container images and dependencies for security weaknesses before deployment.

gastownhall/beads

A project for giving coding agents persistent memory to carry useful context between tasks.

ast-grep/ast-grep

A command-line utility that searches and transforms code by syntax structure rather than plain text alone.

usebruno/bruno

A development environment for exploring APIs and building repeatable API tests.

thedotmack/claude-mem

A memory system that records agent activity, compresses previous sessions and restores relevant context in later sessions.

steel-dev/steel-browser

A browser automation service and sandbox intended for AI agents and applications.

ossf/scorecard

Automated checks that help assess the security practices of open-source projects.

git-town/git-town

Tools for simplifying day-to-day Git branch workflows.

dagger/dagger

An automation engine for building, testing and releasing software across local, CI and cloud environments.

stacklok/toolhive

A platform for operating and administering servers that expose tools through the Model Context Protocol.

traceloop/openllmetry

Adds OpenTelemetry-based monitoring to applications that use language models.

web-infra-dev/midscene

A GUI automation agent aimed at end-to-end interface testing.

modelcontextprotocol/inspector

A visual interface for testing MCP servers and inspecting the tools they provide.

VectifyAI/PageIndex

Indexes documents as a hierarchy so retrieval can use reasoning and tree navigation rather than vector similarity alone.

boxd.sh/downloads/install.sh

A downloadable shell installation script hosted by boxd.sh. This bookmark points to the script rather than a documentation page.

dmtrKovalenko/fframes

A framework for creating videos in Rust and SVG, with GPU rendering and tools for inspecting frames and audio.

Viktor | Not a tool. A hire.

An AI assistant for Slack and Microsoft Teams that connects business tools and performs tasks such as reporting and campaign work.