Skip to content
a16za16z

Marc Andreessen & Amjad Masad on “Good Enough” AI, AGI, and the End of Coding

Amjad Masad, founder and CEO of Replit, joins a16z’s Marc Andreessen and Erik Torenberg to discuss the new world of AI agents, the future of programming, and how software itself is beginning to build software. They trace the history of computing to the rise of AI agents that can now plan, reason, and code for hours without breaking, and explore how Replit is making it possible for anyone to create complex applications in natural language. Amjad explains how RL unlocked reasoning for modern models, why verification loops changed everything, whether LLMs are hitting diminishing returns, and if “good enough” AI might actually block progress toward true general intelligence. 00:00 Intro 00:37 Programming in Plain English 03:00 The Vision Behind Replit 05:15 From Machine Code to English Code 07:00 Building Apps with AI Agents 09:30 When the Agent Becomes the Programmer 11:00 Long-Horizon Reasoning and Coherence 13:45 Reinforcement Learning and Problem Solving 17:30 The Verification Loop and Multi-Agent Systems 21:15 Watching AI Work Like a Human Programmer 23:45 From Stochastic Parrots to Real Reasoning 26:00 Why Coding Is Advancing Faster Than Other Fields 30:15 Verifiable Domains: Math, Code, and Physics 33:45 The AGI Debate: Are We on Track? 37:45 Transfer Learning and the Limits of Human Intelligence 41:15 Functional AGI and Automating Labor 45:20 GPT-5, Diminishing Returns, and Lost “Humanity” 53:10 Creativity, Reasoning, and Finding Truth in AI 57:30 The Origins of Replit and Early Coding Days 01:03:00 Hacking His University and Getting Caught 01:08:00 The Redemption and Lessons Learned for the AI Age Stay Updated: If you enjoyed this episode, be sure to like, subscribe, and share with your friends! Resources: Follow Amjad on X: https://x.com/amasad Follow Marc on X: https://x.com/pmarca Find a16z on X: https://x.com/a16z Find a16z on LinkedIn: https://www.linkedin.com/company/a16z Listen to the a16z Podcast on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYX Listen to the a16z Podcast on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711 Follow our host: https://x.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see a16z.com/disclosures.

Marc AndreessenhostAmjad MasadguestErik Torenberghost
Oct 22, 20251h 11mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Replit’s agents turn English into apps, fueling AGI debates today

  1. Replit’s AI agent workflow lets users describe an idea in plain language and have the system choose a stack, write code, provision infrastructure, test in a browser, and deploy to production in minutes.
  2. The practical bottleneck in democratizing software has shifted from environment setup and tooling to syntax itself, motivating the push from “code” toward “typing thoughts” in natural language.
  3. Agent reliability hinges on long-horizon coherence, which Masad attributes to reinforcement learning plus product-layer techniques like context compression and explicit verification loops.
  4. Coding is advancing faster than most fields because it provides scalable, automated verification (compile/run/unit tests), whereas domains like law and healthcare remain “squishy” and hard to score deterministically.
  5. They debate whether current progress leads to “true AGI” (efficient continual learning and transfer) or a “functional AGI” that automates labor via domain-by-domain data, risking a ‘local maximum’ where systems are economically good enough without becoming fully general.

IDEAS WORTH REMEMBERING

5 ideas

Replit’s core promise is “idea → deployed app” with minimal setup friction.

Users describe what they want in plain language; the agent selects the stack, creates DB and services (e.g., payments), writes and tests code, then publishes to cloud infrastructure with a few clicks.

The agent has become the primary user of the development environment.

Masad notes Replit internally realized the “user” shifted from the human to the agent that edits files, installs packages, provisions databases, and browses the app—changing how performance and latency are experienced globally.

Long-horizon agent performance is now a key measurable product metric.

Replit tracks success via real outcomes (users publishing apps) rather than only benchmarks, and Masad claims internal generations improved from ~2 minutes (Agent 1) to ~20 minutes (Agent 2) to ~200 minutes (Agent 3) on meaningful tasks.

Reinforcement learning unlocks step-by-step problem solving beyond pure next-token prediction.

Masad argues pretraining alone doesn’t reliably produce long reasoning chains, while RL in executable environments (bugs with unit tests/known PRs) rewards successful “trajectories,” teaching models how to reach verifiable solutions.

Verification loops are the scaling trick for multi-hour agent work.

By inserting automated checking (unit tests, browser-based testing, kernel execution), Replit can summarize progress, spawn new trajectories, and use multi-agent handoffs—more like a relay race than a single monolithic attempt.

WORDS WORTH SAVING

5 quotes

Ultimately, English is the programming language.

Amjad Masad

When we did this shift, we hadn't realized internally at Replit how much the actual user stopped being the human user, and it's actually the agent programmer.

Amjad Masad

Agent 1, the agent can run for two minutes... Agent 2 came out in February. It ran for twenty minutes. Agent 3, two hundred minutes.

Amjad Masad

Which is why coding is moving faster than any other domain... is because we can, we, w- we can generate these problems and verify them on the fly.

Amjad Masad

We're dealing with magic here that we, I think probably all would've thought was impossible five years ago-

Marc Andreessen

Programming in plain English (natural language as the interface)Replit’s “agent as programmer” product modelAccidental vs essential complexity in software creationLong-horizon coherence and context compressionReinforcement learning from verifiable tasks (code execution)Verification loops and multi-agent relay workflowsAGI vs “functional AGI,” transfer learning, and diminishing returns

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome