AI Agents

concepts · 101 notes linked

Related: Large Language Models · Agentic Coding · Anthropic · Google · Openai · Retrieval Augmented Generation · Multi Agent Systems · Model Context Protocol

Notes

• $85,000 in tokens later: What I learned from scaling agentic coding at Lovable — Scaling agentic coding workflow lessons at Lovable
• 5 Most Popular Agentic AI Design Patterns Every AI Engineer Should Know — Five foundational agentic AI design patterns for autonomous systems
• A Deep Dive Into MCP and the Future of AI Tooling — MCP as standard interface for AI tool use
• AI #174: You're It — Weekly AI roundup, Claude Tag, medical scanners, agent security
• AI #175: The Fable Continues — Weekly AI roundup covering Fable's return aftermath
• AI 2027 — HN debate on AI 2027 scenario: AGI timelines, LLM limits, alignment risks
• AI 50 2025: AI Agents Move Beyond Chat — Forbes AI 50 marks shift from chatbots to full workflow automation
• AI Agents Are Now Trading IP Rights With Each Other—And Earning Crypto for Their Owners - Decrypt — Blockchain platform enabling AI agents to buy and sell tokenized IP rights
• AI Agents Landscape & Ecosystem (July 2026): Complete Interactive Map — Interactive ecosystem map of AI agents and autonomous tools July 2026
• AI Agents for the Enterprise | StackAI — Enterprise platform orchestrating AI agents with governed agentic workflows
• AI Sentience, Agency and Catastrophic Risk | TWIML - The Voice of Machine Learning & AI — Yoshua Bengio discusses AI catastrophic risk and governance
• AI agent designs a complete RISC-V CPU from a 219-word spec sheet in just 12 hours — Verkor Design Conductor autonomously produces verified RISC-V CPU in 12 hours
• AMD Researchers Introduce Agent Laboratory: An Autonomous LLM-based Framework Capable of Completing the Entire Research Process — Autonomous LLM pipeline completing literature review, experiments, and paper writing
• About — AAAI 2024 workshop on cooperative multi-agent decision-making and learning
• Agent, Know Thyself! (and bid accordingly) — Markets to route AI agent tasks via self-assessment
• Agent-Driven Development in Cursor: Testing, Benchmarking, and Optimizing Functions — Cursor agent writes tests, benchmarks, optimizes a function
• Agent-in-the-Loop: A Data Flywheel for Continuous Improvement in LLM-based Customer Support — Live human-feedback flywheel continuously improving LLM customer support system
• Agentic Design Patterns — 424-page Springer book cataloguing fundamental agentic AI design patterns
• An Honest Review of Google Antigravity — Honest developer review of Google Antigravity agent-first IDE
• Antigravity 3 Hands-On with Security Mode: Turbo, Auto, Off, Plus New Skills — Google Antigravity 3 features customizable Skills, Secure Mode, and tiered access
• Architecting efficient context-aware multi-agent framework for production — Context engineering architecture in Google ADK agents
• At TED AI 2023, experts debate whether we've created "the new electricity — TED AI 2023 first AI-only TED conference debates AGI benefits and risks
• Avi Chawla (@_avichawla) on X — Visual comparison of traditional RAG versus agentic RAG approaches
• Beyond Code Autocomplete — AMD's holistic AI integration across the full software development lifecycle
• Building a C compiler with a team of parallel Claudes — 16 parallel Claude agents autonomously build Rust C compiler
• Building with Gemini 3 in Jules — Gemini 3 Pro integration into Jules autonomous coding agent
• CAMEL-AI | Finding the Scaling Laws of Agents — CAMEL-AI open-source multi-agent framework for research on agent scaling laws
• Claude Code Tasks Are Here (New Update Turns Claude Code ToDos to Tasks) — Claude Code Tasks feature upgrades ephemeral Todos to persistent cross-session tasks
• Claude Fable 5 and Mythos 5: Capabilities — Fable 5 capability review, benchmarks, classifiers, reception
• Claude Skills are awesome, maybe a bigger deal than MCP — Claude Skills as lightweight Markdown-based agent capability extension beating MCP
• Claude for Life Sciences — Anthropic launches Claude specialization for life sciences research workflows
• Claude's Chrome plugin is now available to all paid users - Engadget — Anthropic's Claude Chrome plugin expands beyond Max plan subscribers
• Clawdbot vs Claude Code: One for Coding, One for Everything Else (Don't Get Confused) — Comparing Claude Code terminal coding tool with Clawdbot general assistant
• Cohere Command Models: AI-Powered Solutions for Enterprise — Cohere Command family of enterprise LLMs for agentic and RAG workflows
• Comparing Memory Systems for LLM Agents: Vector, Graph, and Event Logs — Agent memory patterns compared by latency, hit-rate, failure modes
• Could A Y Combinator AI Startup's Voice Agent Transform Productivity? — YC-backed voice AI agent April manages email and calendar via speech
• DS-STAR: A state-of-the-art versatile data science agent — Iterative plan-verify data science agent handling heterogeneous file formats
• DSPy — Python framework replacing prompt engineering with optimizable typed signatures
• Forget SaaS: The future is Services as Software, thanks to AI — AI flips SaaS model — software itself becomes the autonomous service worker
• GitHub - aipotheosis-labs/aci: ACI.dev is the open source tool-calling platform that hooks up 600+ tools into any agentic IDE or custom AI agent through direct function calling or a unified MCP server. The birthplace of VibeOps. — Open-source platform providing 600+ tool integrations for AI agents via MCP
• GitHub - garg-ankush/scipe: SCIPE is a powerful tool for evaluating and diagnosing LLM (Large Language Model) graphs or chains. — Python tool for root-cause diagnosis of failing LLM chain nodes
• Google AI Introduces DS STAR: A Multi Agent Data Science System That Plans, Codes And Verifies End To End Analytics — Multi-agent system converting natural language to Python for heterogeneous data analytics
• Google Publishes Scaling Principles for Agentic Architectures — Predictive regression framework for selecting optimal multi-agent coordination strategies
• Google Researchers Can Create an AI That Thinks a Lot Like You After Just a Two-Hour Interview — Stanford/Google study simulates 1,000 people as LLM agents from interview transcripts
• Hermes Agent Adds Asynchronous Subagents, So Delegated Work No Longer Blocks the Parent Chat — Hermes Agent's async_delegation toolset enables non-blocking parallel subagents
• How I Automate Grading With LangChain And GPT-4 — LangChain GPT-4 pipeline automating bootcamp homework grading
• How I Use Every Claude Code Feature — Practitioner's guide to Claude Code features and workflows
• How to Build An AI Agent with Function Calling and GPT-5 | Towards Data Science — Building a web-search AI agent via GPT-5 function calling
• How to Create Powerful Loops in Claude Code | Towards Data Science — Using /goal command to create self-verifying autonomous coding agent loops
• How to Design a Production-Grade Multi-Agent Communication System Using LangGraph Structured Message Bus, ACP Logging, and Persistent Shared State Architecture — LangGraph tutorial building ACP message bus with Planner-Executor-Validator agents
• How to Use RLMs in Deep Agents — Recursive language models combat context rot via programmatic subagent orchestration
• Hugging Face Just Released SmolAgents: A Smol Library that Enables to Run Powerful AI Agents in a Few Lines of Code — Lightweight Hugging Face library for building AI agents in three lines
• Hugging Face clones OpenAI's Deep Research in 24 hours — Open-source research agent replicates OpenAI Deep Research in one day
• I Tested Oh My Claude Code The Only Agents Swarm Orchestration You Need — Oh My Claude Code adds multi-agent swarm orchestration to Claude Code
• I used Claude to vibe-code my wildly overcomplicated smart home — Non-coder uses Claude Code to configure Home Assistant
• I used Gemini 2.0 to create an AI shopping assistant — it's surprisingly good at saving me time and money — Gemini 2.0 Flash used to build reusable agentic shopping assistant prompt
• IDEcline: How the world's most powerful coding tools became second-class citizens overnight — Three-wave shift from IDE-centric to agent-control-plane software development
• Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversations — Co-STORM multi-agent system for serendipitous unknown-unknown discovery
• Introducing Amazon Bio Discovery | Amazon Web Services — AWS agentic platform for lab-in-the-loop drug discovery
• Keynote at NVIDIA GTC San Jose 2026 — Jensen Huang's 2026 GTC keynote covering full AI stack advances
• LLMs-local — awesome platforms, tools, and resources for running LLMs locally — Awesome list for running LLMs locally
• Llamar.ai: A deep dive into the (in)feasibility of RAG with LLMs — RAG product prototype built and shut down due to GPT-4 API cost infeasibility
• MCP + Google Sheets: A Beginner's Guide to MCP Servers — Beginner tutorial connecting MCP servers to Google Sheets
• MCP doesn't move data. It moves trust — MCP as AI governance and intent-control layer over APIs
• MCP: The Missing Link Between AI Agents and APIs — MCP standardizes how AI agents access external APIs
• Maybe showing off an AI-generated fake TV episode during a writers' strike is a bad idea — Fable Studios demos AI TV showrunner agent during Hollywood writers strike
• Meet DrugAgent: A Multi-Agent Framework for Automating Machine Learning in Drug Discovery — LLM multi-agent system automating end-to-end drug discovery ML pipelines
• Memora scales agent memory to boost long-horizon productivity — Harmonic memory system decoupling storage and retrieval for long-horizon agents
• Meta Superintelligence Labs executives are pushing staff to ditch slow internal systems for faster engineering tools — Meta Superintelligence Labs abandons slow internal infra for Vercel and GitHub
• Meta's Yann LeCun predicts 'new paradigm of AI architectures' within 5 years and 'decade of robotics' | TechCrunch — LeCun forecasts LLM paradigm obsolescence and rise of world models
• Nvidia Wants to Replace Nurses With AI for $9 an Hour — Nvidia-Hippocratic AI partnership deploys generative AI nurses at $9/hr
• OmniThink: A Cognitive Framework for Enhanced Long-Form Article Generation Through Iterative Reflection and Expansion — Iterative reflection framework improving knowledge density in LLM long-form writing
• OpenAI is throwing everything into building a fully automated researcher — OpenAI's North Star automated AI researcher agent
• OpenClaw vs Hermes Agent: Why Nous Research's Self-Improving Agent Now Leads OpenRouter's Global Rankings — Hermes Agent surpasses OpenClaw on OpenRouter with self-improving do-learn-improve loop
• Orchestrate teams of Claude Code sessions - Claude Code Docs — Claude Code agent teams for parallel coordinated work
• PocketFlow: 100-line LLM framework — Minimalist 100-line graph-based LLM agent framework
• Practical Guide on how to build an Agent from scratch with Gemini 3 — Step-by-step Python agent implementation using Gemini 3 tool-use loop
• Prompt Engineering Guide — Primer on LLM-powered agents: capabilities, design patterns, use cases
• RL4HCI — Workshop agenda building RL research agenda for human-computer interaction
• ReAct: Synergizing Reasoning and Acting in Language Models — Interleaved reasoning traces and external actions improve LLM agent reliability
• SCIPE - Systematic Chain Improvement and Problem Evaluation — LangChain-featured tool for identifying failing nodes in LLM chains
• Silicon Valley bets big on 'environments' to train AI agents | TechCrunch — RL environments emerge as critical training infrastructure for capable AI agents
• State of AI, September 2025: A Broad Primer from an AI Team Lead in Education — Education-focused overview of AI history, capabilities, agents, and AGI trajectory
• The AI Agent Era Requires a New Kind of Game Theory — CMU researcher on agent security risks and multi-agent game theory
• The Batch | DeepLearning.AI | AI News & Insights — DeepLearning.AI weekly AI news and insights newsletter homepage
• The Death of the "Everything Prompt": Google's Move Toward Structured AI — Google Interactions API for stateful agentic AI
• The Era of Agentic Organization: Learning to Organize with Language Models — AsyncThink paradigm: concurrent LLM reasoning optimized via reinforcement learning
• The Machine Learning Practitioner's Guide to Agentic AI Systems - MachineLearningMastery.com — Roadmap for ML practitioners transitioning to production agentic systems
• The New Data Commons MCP Server Unlocks a Wealth of Public Datasets for AI Developers — Google Data Commons MCP server enables natural-language access to public datasets
• The transplantable skeleton: Why agentic AI infrastructure must survive corporate surgery — Designing portable agentic AI infrastructure that survives mergers and divestitures
• This startup says it's made ChatGPT for construction sites. Read the pitch deck it used to raise $40 million. — Trunk Tools construction-specific AI agent raises $40M Series B
• Toolformer: Language Models Can Teach Themselves to Use Tools — Self-supervised LLM fine-tuning to autonomously call and integrate external APIs
• Towards Data Science — Benchmark on latency, cost, reproducibility for AI agents
• TradingAgents: Multi-Agents LLM Financial Trading Framework — Multi-agent LLM framework simulating specialized trading firm roles
• What Is Gibberlink Mode, AI's Secret Language? — AI-to-AI protocol enabling machine-efficient communication beyond human language
• When AI Agents Have Their Own Economy, Everything Changes - Decrypt — Crypto infrastructure enables AI agents to participate autonomously in economies
• Why (Senior) Engineers Struggle to Build AI Agents — Senior engineers' deterministic instincts conflict with probabilistic agent design
• Will AI end everything? A guide to guessing | EAG Bay Area 23 — EA Global talk estimating ~19% probability of AI-caused civilizational doom
• Xiaomi's MiMo Code claims it beats Claude Code past 200 steps — Long-horizon coding agent endurance gap and competing harness approaches
• elvis (@omarsar0) on X — Stanford CME295 new course on Transformers and LLMs announced
• get-physics-done — open-source agentic AI physicist (PSI) — Open-source agentic AI system for physics research