The Rise Of AI Coworkers

Y Combinator · 2026-11-14 · 60 min
https://www.youtube.com/watch?v=y1-9yeYLJUoVideo summary
Grok Bot’s Lauren Tan and Roman Ugarte describe AI coworkers, 4,000 monthly PRs, and why clear goals matter as execution gets cheaper.
Grok Bot creators Lauren Tan and Roman Ugarte explain how they built the AI teammate product in seven weeks, giving bots identities, tools, and persistent roles. Teams use them for marketing campaigns, support, sales research, and engineering; specialists can be shared across a company and earn broader access as trust grows. Tan says her engineering workflow now runs in Slack, where a bot helped build a feature end to end in a 200-message thread, while her output has topped 4,000 pull requests a month. She stresses automated checks, codebase guardrails, and agents that run software to verify their own work. The guests also discuss bot collaboration, permissions, and the challenges of understanding workplace authority. As execution becomes faster and cheaper, they argue, the bottleneck shifts to defining goals, aligning on product direction, and applying human judgment rather than shipping every possible feature. Tan also describes moving from an accounting degree to engineering through self-teaching and sharing her work publicly.
Chapters
- 0:00Intro: Grok Bot’s AI Teammates Built in Seven Weeks
- 3:36Giving Bots Real Jobs: Ad Bots Manage Campaigns and Lauren’s Agents Work in Slack
- 6:49One Bot or a Team of Specialists: Customer Bots, Bookkeeping, and Focused AI Teammates
- 10:39Sharing Bots Across the Company: Team Bots Build Sales and Engineering Software Factories
- 14:46Identity and Permissions for AI Coworkers: Role-Based Access and Visible Bot Chats
- 18:37Bot-to-Bot Collaboration and Trust: Potato Consults Data Bot as Access Expands Gradually
- 22:35The Workflow Behind 4,000 PRs a Month: Lauren Builds a Feature with Potato in Slack
- 26:00Teaching Agents to Write Code You Trust: Guardrails, Self-Merging PRs, and Lauren’s Pstack
- 29:01Building a Personal Software Factory: Cappy, GStack, and 10–15 Agent Threads
- 33:44The Future of AI Coworkers: 2027 Identities, Neuralink Telepathy, and Company-Wide Factories
- 36:47Rethinking the AI-Native Company: Three Roles and Bots Triage Hundreds of Customer Messages a Minute
- 40:13Teaching Agents to Check Their Own Work: Runtime Feedback Loops and Cursor Projects
- 43:56How Much Code Review Is Enough? Lauren Favors Codebase Rules Over Token-Heavy Reviews
- 47:33Codebase Guardrails That Save Tokens: Electron Thread Rules and Dependency Checks
- 52:20When Product Judgment Becomes the Bottleneck: 4,000 PRs a Month and Human-Curated Releases
- 56:29From Finance to AI Engineering: Lauren Tan’s Ember Blog Led to a Portland Conference Invitation and Tech Opportunities
What the full analysis includes
16 chapters · 84 key points · 16 section summaries · mind map
Preview of the first chapter · Intro: Grok Bot’s AI Teammates Built in Seven Weeks
The remaining 81 key points, section summaries and mind map are in the full analysis. Analyze this video to see everything →
This video is 1 hours. The Free plan analyzes videos up to 30 min; plans from Plus cover videos up to 2 hours.
This is a Tier 1 public summary
Whether the chapter key points, section summaries and mind map are public is up to the person who shared it. Want the full analysis?Submit one yourself.
More from this channel
Is AI Now Good Enough To Cure Cancer? | YC Paper ClubY CombinatorYC researchers show how AI models trained on DNA and patient data could accelerate personalized cancer treatment and biological design.
Bringing Supersonic Flight To The PeopleY CombinatorBlake Scholl says Rolls-Royce’s 2022 exit pushed Boom to build engines, unlocking quiet supersonic flight and AI power turbines.
How Amjad Masad Saw Vibe Coding A Decade EarlyY CombinatorAmjad Masad explains how Replit’s 2024 coding agent anticipated vibe coding and why banks value its deep infrastructure.
What Happens When Everyone Can Build SoftwareY CombinatorSupabase CEO Paul Copplestone says AI builders could drive 1.2 billion personal databases monthly, as Supabase surpasses 13 million users.
The Case For Data Centers In SpaceY CombinatorStarcloud CEO Philip Johnston says orbital data centers could add 20 GW of compute, with 88,000 satellites planned.
Related analyses
Inside Salesforce’s Acquisition Strategy for the AI Eraa16zSalesforce’s John Somorjai says Fin resolves 65% of support traffic without humans, while Slack revenue has nearly tripled since acquisition.
Circleback CEO Ali Haghani: Why Your Company Should Be Recording More MeetingsY CombinatorCircleBack CEO Ali Haghani says companies should record conversations so AI agents retain context, while humans approve emails and security decisions.
EP335. 蘋果摺疊手機登場、大 Agent 時代來了、Vibe Coding 101 | M觀點M觀點蘋果 iPhone Duo 定價最高 11.89 萬元,Grok Bot、Meta Muse 與微軟 Project Opal 推進雲端 AI Agent。
Building Enduring Companies Outside Silicon Valley | Santiago Suárez of Addia16z Deep DivesAddi CEO Santiago Suárez built a Colombian financial platform serving 3 million consumers, with AI agents resolving nearly 80% of customer queries.
EP160 - AI 時代的公司將徹底改變!一個人如何創造百倍價值?Anthropic、SpaceXAI 已經讓我們看見答案!科技浪Anthropic 用 Scion 將每日程式更新推升至 3 萬次,主持人主張以乾淨環境建立 AI agent 信任,讓小團隊勝過大公司。