Feb 17, 2026 · 50m · y-combinator
Inside Claude Code With Its Creator Boris Cherny · Y Combinator
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
Boris Cherny, Creator and Head of Claude Code at Anthropic, joins Y Combinator partners on The Light Cone podcast to discuss the genesis, design philosophy, and future of AI-native developer tools. He outlines Anthropic's core doctrine of building for future model capabilities, the explosive growth of autonomous agent workflows, and how LLMs are fundamentally redefining modern software engineering.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
The partners as informed peer 4.3 Guest teaching 5.1 Guest disagreement 1.0 The partners pushing back 0.1
speaking balance: gold is the partners, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 25:03 Plan mode lifespan pushback
Boris directly challenges the premise that plan mode prompt scaffolding will be needed long-term, bluntly predicting its obsolescence within a single month.
Hardest push from the partners ▶ 26:47 Challenging model autonomy without human planningThe co-host pushes back on whether human conceptualization can really be replaced, questioning if developers will always need a designated planning step.
Biggest teaching moment ▶ 10:01 Delete your bloated CLAUDE.mdBoris corrects common user assumptions about prompt files, instructing the hosts to delete multi-thousand-token prompt files and rely on minimalist instructions.
The partners hold their own ▶ 17:32 YC technical hiring framework via Claude CodeGarry Tan demonstrates his operator depth by explaining YC's live evaluation framework that analyzes candidate Claude Code transcripts across test coverage and architectural planning.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The partners as informed peer | Guest teaching | Guest disagreement | The partners pushing back | Why |
|---|---|---|---|---|---|---|
| Welcome and Initial Reactions to Claude Code | 4 | 5 | 1 | 0 | The hosts open enthusiastically about their obsession with Claude Code, and Boris explains the founding philosophy of building for models six months out rather than current capabilities. | |
| Choosing the CLI Form Factor and Exploration Mode | 3 | 4 | 1 | 0 | The co-host queries the contrarian decision to use a CLI form factor when full IDEs were the trend, and Boris shares how it originated as a quick zero-UI prototype that rapidly gained internal traction. | |
| Early Use Cases, CLAUDE.md Origin, and Scaffolding | 4 | 6 | 2 | 0 | Boris explains the origins of CLAUDE.md from latent demand and educates the host on why his personal configuration is only two lines, advising users to delete bloated files rather than over-engineering scaffolding. | |
| Dev Tool Diversity and Terminal Accessibility | 5 | 4 | 1 | 1 | The host provides specific opinions on terminal verbosity and stopping runaway bug loops, while Boris explains the balance between internal power users wanting raw bash output and mainstream usability. | |
| AI-Assisted Debugging and The Memory Leak Story | 6 | 5 | 1 | 0 | The host discusses modern bug-fixing workflows with Sentry and MCP, while Boris relates how an engineer solved a memory leak using Claude Code faster than his own manual heap dump analysis. | |
| Hiring for First Principles, Scientific Mindsets, and Humility | 6 | 4 | 1 | 0 | Boris advocates for humility and scientific first principles over senior dogma, leading the host to share YC's practice of reviewing Claude Code interaction transcripts to evaluate engineering talent. | |
| Engineer Archetypes and Autonomous AI Workflows | 5 | 5 | 1 | 0 | Boris describes the bimodal split between hyper-specialists and extreme generalists, emphasizing workflows where engineers use agents to write testing tools and automate the entire SDLC. | |
| Claude Teams, Agent Topologies, and Subagents | 4 | 6 | 1 | 0 | Boris breaks down agent topologies, uncorrelated context windows as test-time compute, and how the plugins system was autonomously built over a single weekend by a swarm of subagents. | |
| The Evolution and Future Lifespan of Plan Mode | 4 | 6 | 2 | 1 | Boris delivers a surprising prediction that plan mode will be obsolete within a month, explaining how improved model intelligence progressively eliminates human babysitting. | |
| Product Philosophy: Latent Demand and Form Factors | 3 | 5 | 1 | 0 | Boris explains the product doctrine of latent demand, describing how observing unprompted user behaviors in the office and online shaped Claude Code's roadmap across diverse client form factors. | |
| Advice for DevTool Startups: Serving Model Demand | 4 | 5 | 1 | 0 | The conversation shifts to TypeScript's design philosophy versus academic languages, with Boris and the co-host agreeing on the necessity of meeting developers where they actually code. | |
| Crafting Terminal UIs and Rapid AI Prototyping | 4 | 4 | 0 | 0 | Boris discusses the technical challenges of React-driven terminal UI design, ANSI escape codes, and rapid prototyping workflows that replaced traditional design tools. | |
| Core Tenets: The Bitter Lesson and Codebase Lifespans | 4 | 6 | 2 | 0 | Boris references Rich Sutton's The Bitter Lesson, arguing that product scaffolding is temporary technical debt that is systematically erased by next-generation base models. | |
| Unprecedented Productivity Gains at Anthropic | 5 | 5 | 0 | 0 | The host highlights claims of 1000x developer leverage, and Boris shares internal metrics showing a 150% productivity surge per engineer contrasted against his past Meta infrastructure experience. | |
| Predictions for 2025: From Engineers to Builders and Safety Milestones | 3 | 6 | 1 | 0 | Boris lays out 2025 forecasts including the elimination of manual IDE coding, the evolution of SWEs into broad builders, and the critical threshold of ASL-4 recursive self-improvement safeguards. | |
| Claude Code's Global Adoption and NASA Mars Rover Impact | 4 | 5 | 0 | 0 | Boris outlines Claude Code's expansion across non-technical domains like data recovery and Mars Rover path plotting, before the host closes with praise for the tooling's impact. |
Statements from this episode (34)
Cherny: Claude Code wrote only 10 percent of his code at launch
“Even in February when we GA'd it wrote maybe like 10% of my code or something like that. I didn't really use it to write code. It wasn't very good at it. I still wrote most of my code by hand.”
Anthropic designs tools for AI models expected six months in the future
“At Anthropic, the way that we thought about it is we don't build for the model of today. We build for the model six months from now.”
Anthropic's path to safe AGI centers on coding and tool use
“You know, as Anthropic, I think for Ant, the bet has been coding
For a long time.
And the bet has been the path to save, to safe AGI is through coding.
And this is, this has kind of always been the idea.
And the way you get there is you teach the model how to …”
Cherny built Claude Code in the terminal to avoid building a UI
“Yes, because I didn't have to build a UI. So it was just me.”
Claude 3.5 Sonnet wrote AppleScript unprompted to detect Cherny's playing music
“And I asked it, what music am I listening to? He wrote some, like, Apple script to script my Mac and look up the music in my music player. And this was sauna 3.5.”
Scaffolding gains get wiped out by next-generation AI models
“There's this other general principle that I think is maybe interesting where you can build for the model and then you can build scaffolding around the model in order to improve performance a little bit. And depending on the domain, you can improve performance,…”
Claude Code stayed a CLI because graphical UIs obsolete too quickly
“And really, I think that's why we stayed in the CLI is because we felt there is no UI we could build that would still be relevant in six months because the model was improving so quickly.”
Delete oversized CLAUDE.md files and start fresh rather than compacting them
“If you hit this, my recommendation would be delete your QuadMD and just start fresh.”
Anthropic employees revolted when Claude Code hid bash output
“I remember early on, this is maybe six months ago, I tried to get rid of bash output, just internally, just to, like, summarize it, because I was like, these giant long bash commands, I don't actually care. And then I gave it to anthropic employees for a day, …”
Claude Code autonomously wrote a tool to debug a memory leak
“And quad code, like took the heap dump. It wrote a little tool for itself to like analyze the heap dump. And then it found the leak faster than I did.”
Cherny: AI makes traditional senior engineering opinions less relevant
“In my old job at a big company, when I hired like architects and this kind of a type of engineer, you look for people that have a lot of experience and really strong opinions, but it actually turns out a lot of this stuff just isn't relevant anymore. And a lot…”
Cherny: The most effective AI engineers are hyper-specialists or hyper-generalists
“When I look at engineers on the team that I think are the most effective, there's essentially two, it's very bimodal. There's one side where it's extreme specialists and so, like, I named Jared before, like, he's a really good example of this, and kind of the …”
Anthropic uses Claude Agent SDK to automate reviews and deployments
“We use the Quad Agent SDK to automate pretty much every part of development. It automates code review, security review it labels all of our issues. It shepherds things to production. It does pretty much everything for us.”
Claude Code's plugin feature was built entirely by an autonomous swarm
“I think the first kind of big example where it worked is our plugins feature was entirely built by a swarm over, over a weekend. It just ran for like a few days. There wasn't really human intervention and plugins is pretty much in the form that it was when it …”
Cherny believes most agents today are prompted as Claude subagents
“I haven't pulled the data on this, but I would bet the majority of agents are actually prompted by quad today in the form of subagents.”
Cherny predicts plan mode in AI tools has a limited lifespan
“I think plan mode probably has a limited lifespan.”
Cherny claims recent Claude Opus models execute plans without babysitting
“And nowadays what I find with Opus 4.5, I think it started with 4.6, it got really good. Once the plan is good, it just stays on track, and it'll just do the thing exactly right almost every time.”
Claude Code autonomously messages Anthropic engineers on Slack via git blame
“A really common pattern is I ask claw to build something. It'll look in the code base. It'll see some engineer touch something in the git flame and then it'll message that engineer on Slack. Just like asking a clarifying question. And then once it gets the ans…”
Cherny predicts Claude will increasingly discover product ideas by analyzing logs
“And I think Quad is going to get increasingly good at kind of figuring out these kind of product ideas for you, just because it can look at feedback, it can look at debug logs, it can kind of figure this out.”
Cherny: Dev tool startups should enable what AI models naturally attempt
“The way I would frame it is think about the thing that the model wants to do and figure out how do you make that easier? And that's something that we saw, you know, like when I first started hacking on Claude code, I realized like this thing just wants to use …”
Claude Code enables building 20 UI prototypes in a few hours
“You can write these prototypes and you can just do like 20 prototypes back to back, see which one you like, and then ship that, and the whole thing takes maybe a couple hours. Whereas in the past, what you would have had to do is like learn to use origami or f…”
No code in the Claude Code codebase is over six months old
“All of CloudCode has just been written and rewritten and rewritten and rewritten over and over and over. We unship tools every couple weeks. We add new tools every couple weeks. There's no part of CloudCode that was around six months ago. It's just constantly …”
All technical staff and half of Anthropic sales use Claude Code
“Internally, if you look at like technical employees, they all use quad code every day. And even non-technical employees, I think like half the sales team uses quad code.”
Claude Code increased Anthropic engineer productivity by 150 percent
“And since Quad Code came out, productivity per engineer at Anthropic has grown a 150%.”
Cherny: Anthropic operates as a research lab prioritizing models over product
“So one is it operates as a research lab. So the product org was teeny, teeny, tiny. It's really all about building a safe model. That's all that matters. And so this idea of just being very close to the model and being very close to development and being not t…”
Cherny uninstalled his IDE and lands 20 AI-written PRs daily
“For me personally, it's been a hundred percent for like since Opus 4.5. I just, I uninstalled my IDE. I don't edit a single line of code by hand. It's just a hundred percent quad code in Opus. And you know, I land, you know, like 20 pairs a day, every day.”
Claude generates between 70 and 90 percent of Anthropic's codebase
“If you look at Anthropic overall, it ranges between, like, 70 to 90% you know, depending on the team. For a lot of teams, it's also, like, a hundred percent. For a lot of people, it's a hundred percent.”
Cherny predicts AI will generally solve coding for everyone
“So continuing to trace the exponential, I think what will happen is coding will be generally solved for everyone. And I think today coding is practically solved, you know, for me, and I think it'll be the case for everyone you know, regardless of domain.”
Cherny: The software engineer title will disappear as roles become builders
“I think we're gonna start to see the title software engineer go away, and I think it's just gonna be maybe builder, maybe product manager, maybe we'll keep the title as kind of a vestigial thing, but the work that people do, it's not just gonna be coding, it's…”
Non-engineering roles at Anthropic already write code, a trend spreading everywhere
“Every single function on our team codes, like our PM's code, our designers code, our EM codes, our like everyone, our finance guy codes, like everyone on our team codes. We're going to start to see this everywhere.”
Mercury data shows 70 percent of startups choose Claude
“There was some stuff from Mercury that like, 70% of startups are, you know, choosing quad as their model of choice.”
SemiAnalysis reports Claude Code generates four percent of all public commits
“There were some other stuff from like semi-analysis that four percent of all public commits are made by quad code.”