LLM Wiki — Synaptic Lattice

Sequoia Ascent 2026 for Founders

Properties1
tagsvibe-coding, founder, software-3.0, agentic-engineering, jagged-intelligence, verifiability

Leitura de 5 min

Sequoia Ascent 2026: Karpathy’s Fireside Chat for Founders

Andrej Karpathy’s talk at Sequoia Ascent 2026 offers valuable insights for non-technical founders building AI-native companies. Here are the key takeaways relevant to the vibe coding and founder experience:

The Agentic Inflection Point (December 2025)

Around December 2025, Karpathy noticed a step change in AI agent reliability. Generated code chunks became larger, more coherent, and more reliable. The unit of programming shifted from typing lines of code to delegating larger “macro actions.” This is when founders can start trusting agents with more work.

For founders: The default workflow changed. Tools like Claude Code, Codex, and Cursor became useful but required frequent correction. Around December 2025, the step change meant founders could start delegating more.

Software 3.0: The Context Window as the New Program

Software evolution:

  • Software 1.0: Humans write explicit code
  • Software 2.0: Humans create datasets, objectives, neural networks; program learned into weights
  • Software 3.0: Humans program LLMs through prompts, context, tools, examples, memory, and instructions. The context window becomes the main lever.

For founders: In Software 3.0, the installer can be a block of instructions an agent reads and adapts to the local environment. Much more adaptive than brittle shell scripts. This means founders can have agents set up development environments, install tools, and configure infrastructure through natural language prompts.

Vibe Coding vs. Agentic Engineering

  • Vibe coding: Raises the floor. Lets almost anyone create software by describing what they want. Fine for prototypes and personal tools.
  • Agentic engineering: Raises the ceiling. Professional discipline of coordinating fallible agents while preserving correctness, security, taste, and maintainability.

For founders: Vibe coding is fine for prototypes and personal tools. But serious teams need agentic engineering — the discipline of coordinating agents while preserving quality. Hiring should test this directly: build a substantial project with agents, deploy it, have adversarial agents try to break it.

Hiring for the Agentic Era

Traditional coding puzzles are mismatched for the new era. Better interview approach:

  • Give candidates a big project and see them implement it
  • Have agents simulate activity on the deployed project
  • Test: Can the candidate decompose work for agents? Write useful specs? Preserve quality while moving fast? Review generated work? Secure and harden the system? Use agents as leverage?

For founders: If agentic engineering is the new professional skill, hiring should test it directly. Don’t rely on traditional coding puzzles.

Agent-Native Infrastructure

Most software is still built for humans clicking through screens. Agent-native surfaces needed: Markdown docs, CLIs, APIs, MCP servers, structured logs, machine-readable schemas, copy-pasteable agent instructions, safe permissioning, auditable actions, headless setup flows.

For founders: Products need agent-native surfaces first. Don’t tell humans “go to this URL” or “click here” — describe things to agents first, build automation around data structures legible to LLMs.

Education: You Can Outsource Thinking, But Not Understanding

Key line: “You can outsource your thinking, but you can’t outsource your understanding.” Even if agents do more work, the human still needs understanding to direct them. You need to know what is worth building, what question matters, what result is suspicious, and what tradeoff is acceptable.

LLM knowledge bases are tools for transforming information into understanding. The human expert contributes the distilled artifact and the taste behind it; the agent can then explain it interactively to each learner.

For founders: LLM knowledge bases are tools for transforming information into understanding. They enhance both human and agent insight.

Main Thesis for Founders

AI is becoming a new operating layer for digital work. The scarce thing is shifting:

  • Less scarce: code generation, API recall, boilerplate, first drafts
  • More scarce: understanding, taste, eval design, security, system boundaries, agent orchestration, domain-specific feedback loops

For founders, the most important questions:

  • What becomes possible when the primary user is an agent acting for a human?
  • What workflows can be rebuilt around sensors, actuators, and verifiable loops?
  • What software should disappear into direct model transformations?
  • What domains are valuable and verifiable but not yet heavily trained by frontier labs?
  • What human judgment must remain in the loop to preserve quality?

Implications for Founder/PM Non-Technical

  1. Software reorganization: Work is being reorganized around agents. Knowledge work becomes: define context → define tools → define feedback loop → define guardrails → let agents work → preserve human understanding.

  2. Valuable verifiable domains: Create domain-specific environments where models can try actions and receive reliable rewards. Even if not heavily trained by frontier labs, you can use fine-tuning or your own RL environments.

  3. Human judgment remains: Taste, engineering judgment, design, and oversight remain uniquely human. Agents fill in the blanks, but humans design the spec and plan.

  4. Agent-native infrastructure: Products need agent-native surfaces first. The future: agents using sensors and actuators on behalf of people and organizations.

  5. Education enhancement: Tools that enhance understanding are incredibly interesting. LLM knowledge bases help both humans and agents gain insight through synthetic data generation over fixed data.

  • vibe-coding — the pattern of directing AI agents to write software
  • non-technical-founder — the persona that benefits most from these capabilities
  • pm — the product role that orchestrates delivery
  • agent-loop — the while loop pattern underlying all agent architectures
  • context-engineering — crafting the full information environment an agent operates in

External References

  • Full transcript: Sequoia Ascent 2026 fireside chat with Stephanie Zhan
  • Key concepts: Software 3.0, Agentic Engineering, Jagged Intelligence, Verifiability
  • Related: LLM Wiki pattern, context window as program, agent-native infrastructure

Sources

^[raw/external/docs-getsemantica-ai-c0a0b760.md]

Visão de gráfico

Backlinks