People to Watch
The researchers, builders, and educators shaping the autonomous agent ecosystem. Not a ranked list — a reference of voices worth following.
53 of 53 voices
Must-Follow
Known for: Leading AutoGen and creating Magentic-One multi-agent system
Leads AutoGen and Magentic-One at Microsoft Research. Builds and studies how AI agents collaborate with humans on complex tasks. Announced AutoGen 0.4, a full reimagining of the mu…
WHY TO WATCH
Shapes Microsoft's multi-agent framework used by thousands of enterprise developers
Known for: Building Replit and pivoting it into an AI-native coding agent platform
Founded Replit, the online coding platform that pivoted to AI-first development with autonomous coding agents. Replit's agent can run for three hours doing useful work autonomously…
WHY TO WATCH
Proving that autonomous coding agents can replace traditional development environments
Known for: Co-founding Anthropic and co-inventing RLHF
Co-founded Anthropic after serving as VP of Research at OpenAI. Co-inventor of RLHF. Led development of the Claude model family. Drives responsible scaling policy and authored the …
WHY TO WATCH
Shapes AI safety standards and responsible scaling policy across the entire industry
Known for: Co-creating Model Context Protocol (MCP), now an industry standard
Co-created the Model Context Protocol (MCP) at Anthropic, now adopted by OpenAI, Google, and the broader agent ecosystem. MCP addresses the M×N problem of connecting AI models with…
WHY TO WATCH
MCP is becoming the standard for agent-tool connectivity across the ecosystem
Known for: Creating Next.js, Vercel, and v0 AI development tool
Founded Vercel and created Next.js and Socket.IO. Built v0, an AI web development tool that creates applications from natural language. Released the AI SDK for building conversatio…
WHY TO WATCH
Building the deployment infrastructure and AI SDK that agent-built applications run on
Known for: Co-founding OpenAI and launching SSI at $32B valuation
Co-founded OpenAI and served as Chief Scientist before departing to start Safe Superintelligence Inc. (SSI). SSI raised $2B at a $32B valuation with a single mission: build safe su…
WHY TO WATCH
Pursuing the most ambitious AI safety research agenda with massive funding and talent
Known for: Discovering neural scaling laws and co-founding Anthropic
Theoretical physicist turned AI researcher. Co-authored the landmark 'Scaling Laws for Neural Language Models' paper at OpenAI. Co-founded Anthropic where he serves as Chief Scienc…
WHY TO WATCH
His scaling laws research underpins every major AI lab's training strategy
Known for: Building Cursor, the fastest-growing AI code editor
Co-founded Cursor at age 23, which became one of the fastest-growing SaaS products in history — crossing $500M ARR by mid-2025 with 1M+ daily users. MIT CS and Math graduate. Curso…
WHY TO WATCH
Defining what the agentic software development experience looks like for millions of developers
Known for: Creating Devin, the first fully autonomous AI software engineer
Co-founded Cognition AI, creator of Devin — the first autonomous AI software engineer. Won three IOI gold medals. Devin grew from $1M to $73M ARR in nine months. Cognition reached …
WHY TO WATCH
Leading the most ambitious attempt to fully automate software engineering with AI agents
Known for: Creating the ReAct agent reasoning framework (ICLR 2023 oral)
Created ReAct (Reasoning + Acting), the foundational paradigm for LLM agents that interleave reasoning traces with tool use. Also co-authored Tree of Thoughts and SWE-agent. Built …
WHY TO WATCH
Invented the most widely used agent architecture pattern and continues advancing agent research
Known for: Every Level of Claude Code Skills in 27 mins — 27 concepts demonstrated live with timestamped progression from basics to multi-agent orchestration
Claude Code tutorial creator who mapped the entire capability surface — from CLAUDE.md to multi-agent teams — into dense, timestamped, no-hype technical content. Featured in expert…
WHY TO WATCH
One of the few AI content creators who shows real running systems without selling courses. Demonstra…
Former Director of AI at Tesla and founding member of OpenAI. Now independent — building educational AI content and exploring the implications of AI-native software. His writing on…
Built LangChain — one of the most widely adopted LLM application frameworks — and LangSmith for production observability. Thinks deeply about the transition from chains to graphs t…
Writes the most comprehensive technical blog in AI — long-form deep dives on LLM agents, safety, RLHF, and diffusion models. Her post 'LLM Powered Autonomous Agents' (2023) remains…
Creator of Datasette and Django co-creator. Runs one of the most consistently useful blogs on AI tooling — LLM-powered CLI tools, prompt engineering patterns, and practical evaluat…
Runs Latent Space, the leading AI engineering podcast. Editor of The Illustrated Transformer and long-form essays on the AI engineer role. Coined the term 'AI Engineer' and has bee…
Rising Voices
Known for: Co-creating GitHub Copilot, the first mainstream AI coding assistant
Co-created GitHub Copilot, the first widely-adopted AI coding assistant. Serial builder and 2x CEO. Now at Perplexity working on AI-native interfaces. Foundational figure in AI-ass…
WHY TO WATCH
Pioneered the AI coding assistant paradigm that enabled today's agentic code editors
Known for: Leading Claude's character and personality alignment training
Leads Anthropic's personality alignment team, responsible for Claude's character training. PhD in philosophy from NYU. Previously worked on AI safety via debate at OpenAI.
WHY TO WATCH
Defines how the most capable AI agents develop values, honesty, and behavioral character
Known for: Creating GPT Researcher, the top open-source deep research agent
Created GPT Researcher, the leading open-source autonomous research agent with 14K+ GitHub stars. Also building Tavily, a search API optimized for AI agents. Pioneered the planner-…
WHY TO WATCH
Building both the research agent and the search infrastructure agents need
Known for: Building Agent.ai marketplace to 250K+ users alongside HubSpot CTO role
Co-founded HubSpot (NYSE: HUBS). Built Agent.ai as a side project — an AI agent marketplace that surpassed 250K users within months. Publishes the simple.ai newsletter to 2M+ subsc…
WHY TO WATCH
Demonstrating how AI agent marketplaces can create a new platform economy
Known for: Building Lindy AI, the leading no-code AI agent platform
Founded Lindy AI, a no-code platform for building AI agents that automate business workflows across 1,600+ apps. Raised $50M. Previously Head of Product at Uber.
WHY TO WATCH
Making autonomous agents accessible to non-technical business users at scale
Known for: Creating OpenHands coding agent and SWE-Gym benchmark
Professor at Carnegie Mellon and Chief Scientist at All Hands AI. Co-created OpenHands, an open-source coding agent with 31K GitHub stars. Leads research on autonomous software dev…
WHY TO WATCH
Bridges academic research and production-grade open-source coding agents
Known for: Creating Instructor, the standard library for structured LLM outputs
Created Instructor, the most popular Python library for structured LLM outputs with 6M+ monthly downloads and 12K+ GitHub stars. Cited by OpenAI as inspiration for their structured…
WHY TO WATCH
His library is a dependency in most production agent systems that need reliable outputs
Known for: Leading developer experience at OpenAI and now Google AI Studio
Leads product for Google AI Studio and the Gemini API at Google DeepMind. Previously head of developer relations at OpenAI where he helped launch the ChatGPT API era. Bridges AI re…
WHY TO WATCH
Shapes how millions of developers access and build with Google's AI models and agent APIs
Known for: RLHF research, OLMo/Tulu models, and interconnects.ai newsletter
Leads post-training at Ai2, building open models including Tulu and OLMo. Author of 'The RLHF Book'. Writes the widely-read interconnects.ai Substack on RLHF and alignment. Previou…
WHY TO WATCH
Leading voice on open-source post-training methods that make agents more capable
Known for: Creating HELM benchmark and directing Stanford CRFM
Director of Stanford's Center for Research on Foundation Models (CRFM). Created HELM, the most comprehensive holistic evaluation framework for language models. Advocates for transp…
WHY TO WATCH
Sets the standard for how foundation models and agents are evaluated and compared
Known for: Creating Mastra TypeScript agent framework and co-founding Gatsby.js
Co-founded Mastra, the TypeScript AI agent framework, after co-founding Gatsby.js. YC-backed, raised $13M seed. Authored 'Principles of Building AI Agents'. Brings JavaScript ecosy…
WHY TO WATCH
Making AI agents accessible to the massive JavaScript/TypeScript developer ecosystem
Known for: Co-creating Llama models and Meta Agent Research Environment
Co-creator of the Llama model family at Meta. Led the open-sourcing of Meta Agent Research Environment (ARE) and GAIA-2 benchmarks. Research spans RLHF, scaling laws, synthetic dat…
WHY TO WATCH
Drives Meta's open-source LLM strategy and agent evaluation infrastructure
Known for: Creating AutoGPT, one of the most starred GitHub repos ever
Created AutoGPT in March 2023, one of the most starred repositories in GitHub history and the first viral autonomous AI agent. Demonstrated that LLMs could self-direct through mult…
WHY TO WATCH
Sparked the autonomous agent movement and continues pushing open-source agent capabilities
Known for: Building Windsurf/Codeium to 1M+ developers and pioneering agentic IDE features
Co-founded Codeium (rebranded to Windsurf), an AI-powered IDE with 1M+ developers. Launched Windsurf Editor in November 2024. Built one of the fastest-scaling developer tools in hi…
WHY TO WATCH
Built one of the top three agentic code editors competing with Cursor and Copilot
Known for: Building E2B, the go-to sandbox for AI agent code execution
Co-founded E2B, the open-source secure sandbox runtime for AI-generated code execution. Raised $35M from Insight Partners. E2B powers code execution for Manus, Cursor, and other ma…
WHY TO WATCH
E2B is critical infrastructure that powers code execution for top agent platforms
Also Follow
Runs Anthropic's developer relations and is one of the clearest voices on practical prompt engineering and tool-use patterns with Claude. Maintains the Anthropic prompt library and…
Author of 'System Design Interview' series and the ByteByteGo newsletter. Increasingly covering AI system design — RAG architectures, agent infrastructure patterns, and how enterpr…
Built Codeium (now Windsurf) — one of the top AI coding tools and agentic IDEs. Among the most thoughtful voices on what 'agentic coding' actually means operationally: when should …
Built Modal — the preferred deployment platform for AI workloads and agent infrastructure among the builder community. Modal's sandboxed code execution is widely used for giving ag…
Author of 'Designing Machine Learning Systems' and the free AI Engineering book. Teaches ML systems at Stanford and writes deeply about the operational reality of running ML in pro…
Known for: The ACE Framework and pioneering YouTube content on autonomous agents
Full-time AI communicator and researcher. Created the ACE (Autonomous Cognitive Entity) Framework and OpenAI Agent Swarm. Runs a popular YouTube channel documenting agent experimen…
WHY TO WATCH
One of the earliest and most consistent voices documenting autonomous agent development
Known for: The most popular agentic AI engineering course on Udemy (400K+ students)
Co-founder and CTO of Nebula.io. Creates top-rated agentic AI courses on Udemy with 400K+ students across 190 countries. Teaches building autonomous agents, MCP integrations, and p…
WHY TO WATCH
Training the next generation of AI agent developers with hands-on production courses
Wharton professor who runs rigorous experiments on AI adoption and writes accessibly about what actually works. His book 'Co-Intelligence' and Substack 'One Useful Thing' document …
Known for: Co-creating the definitive prompt engineering course with Andrew Ng
Member of technical staff at OpenAI working on post-training research. Contributed to ChatGPT Agent and Deep Research features. Co-created the popular 'ChatGPT Prompt Engineering f…
WHY TO WATCH
Building OpenAI's agent and deep research capabilities from inside the lab
Led the research on Chain-of-Thought prompting at Google Brain, one of the most cited papers in modern AI. Now at OpenAI. His work on reasoning emergence and scaling has directly s…
Runs product at Hugging Face — the central hub for open-source AI models, datasets, and Spaces. Key voice on the open vs. closed model debate and the practical economics of running…
Built LlamaIndex (formerly GPT Index) — the most widely used framework for connecting LLMs to custom data. Think deeply about the retrieval-augmented generation design space, chunk…
Known for: Creating Langfuse, the standard open-source LLM observability platform
Co-founded Langfuse, the leading open-source LLM observability platform (YC W23). Built for LLM tracing, evals, and prompt management. Previously at Google and Cherry Ventures.
WHY TO WATCH
Built the observability layer that most agent developers use to debug and evaluate
Known for: Founding Fixie.ai and bringing academic AI systems expertise to enterprise
Former Harvard CS professor turned AI entrepreneur. Co-founded Fixie.ai for enterprise agent development, then joined Palantir as Head of AI Systems. Publishes on the operational r…
WHY TO WATCH
Bridges academic AI research and enterprise-grade agent deployment at Palantir scale
Known for: AI automation tutorials and building $300K/mo solo creator business
Runs one of the largest AI automation YouTube channels. Founded LeftClick (AI consulting) and Maker School (top no-code community). Creates daily long-form content on n8n and AI ag…
WHY TO WATCH
Demonstrates practical AI agent automation for real business use cases daily
Known for: Creating Reflexion, the verbal reinforcement learning framework for agents
Lead author of Reflexion (NeurIPS 2023), which introduced verbal self-reflection as a reinforcement learning mechanism for language agents. Achieved 91% pass@1 on HumanEval, surpas…
WHY TO WATCH
Reflexion is a foundational technique used in production agent systems for self-improvement
Built Aider — one of the most capable open-source AI coding assistants. Regularly publishes benchmarks comparing AI coding tools across models and publishes detailed analysis of wh…
Known for: Popularizing vibe coding and building VibeCode AI app builder
AI educator with 1M+ followers across platforms. Founded VibeCode, an AI-powered no-code app builder. Popularized 'vibe coding' — using AI to build apps from natural language descr…
WHY TO WATCH
Making AI agent development accessible to non-programmers through the vibe coding movement
Produces some of the clearest technical video content on LLM agents, RAG systems, and applied ML. Bridges the gap between research papers and practical implementation — his impleme…
Google Machine Learning Developer Expert. Runs one of the most practical YouTube channels on LangChain, agent patterns, and Google AI tools. Regular implementations of new agent fr…
Known for: Creating Guardrails AI, the open-source LLM validation framework
Co-founded Guardrails AI, the open-source platform for defining and enforcing AI assurance in generative AI applications. Raised $7.5M seed. Previously Senior ML Engineer at Apple …
WHY TO WATCH
Building the safety layer that production agent systems need for reliable outputs
Created BabyAGI — one of the earliest and most influential autonomous task-execution agents. Partner at Untapped Capital. Runs early-stage experiments in agent architectures before…