AI is making organizations flatter

3 trends in startups/tech/venture

📈 Engineers are increasingly becoming resource allocators

• Historically, the division of labor in technology was clear: engineers built, and business operations optimized. In the AI era, that line has vanished.

• Today, engineers aren't just builders; they are optimizers. Startups can no longer afford to focus purely on coding now and optimizing later.

• Because AI demands massive amounts of compute — processing power, memory, networking, and storage — engineers must weave resource and process optimization into the very first lines of code they write.

🤖 AI is making organizations flatter

• Jensen Huang (CEO of Nvidia, the leading AI company and the most valuable company in the world) recently shared that he has a whopping 60 people directly reporting to him. His rationale? The era of AI requires a more coordinated effort — which requires breaking down barriers to communication.

• Nvidia is not alone in keeping as flat of a hierarchy as possible. In fact, both OpenAI and Anthropic now have a record number of executives who are individual contributors — a phenomenon that’s nearly unheard of among old guard companies.

• I had the opportunity to discuss AI’s impact on organizational structures with KC, OpenAI’s Chief People Officer — you can find the video of our conversation here 👇🏼

💰 Investors are penalizing companies that blame AI for layoffs

• Recent data shows that 56% of public companies citing AI as their primary layoff driver traded in the red post-announcement, with their stocks tumbling an average of 25%.

• This trend extends beyond public markets. A Gartner survey of 350 executives found that firms cutting heads to make room for AI failed to generate higher returns than those that maintained their headcount.

• The clear winners? Companies using AI for “people amplification”—boosting employee productivity rather than replacing humans outright.

2 theses on what’s next

🤖 3 types of jobs are (still) safe from AI (for now)

• As more tasks and workflows are getting assigned to AI agents over humans, 3 key roles are emerging as ones that still remain critical for a human to take on:

  • 1. Builders and systems architects (i.e., people deciding what to build and how to build it with judgment and taste)

  • 2. Relationship experts (i.e., people who know people who can get things done)

  • 3. Validators and context carriers (i.e., people who know why things are the way they are)

🤔 Startups need a minimum viable number of employees to be resilient

• As more companies are wrestling with the optimal ratio of AI agents to people, another critical number is getting overlooked: the minimal number of humans for an organization to be resilient.

• Consider two startups: Startup A and Startup B. Startup A employs 10 people. Startup B employs 3 people who in turn run 10 autonomous agents. When 1 person leaves Startup A, that’s a 10% headcount loss. When 1 person leaves Startup B, that’s a 33% headcount loss. While Startup A has 9 people left to absorb the shock, Startup B only has 2 people left, leading to a serious loss of institutional memory and relationships.

• So, what’s the magic number of employees that's considered “minimally viable” in the age of AI? We’re about to find out.

1 tool I love

🎓 AI assistant for therapists

• Doctors have assistants. Lawyers have paralegals. Why do therapists work alone? Therapists do some of the most emotionally taxing work on the planet, yet they are often the least supported when it comes to the "heavy lifting" of administration and documentation.

• I see this firsthand serving as a voting Commissioner on California's Mental Health Commission. In fact, mental health is a global problem. Every therapist, in every country, runs into the same wall: they're alone with the work no one trained them for.

• Klarify (part of our advise-to-invest program) uses AI to empower therapists, not replace them. Klarify handles every job a therapist does other than the therapy itself: clinical notes, treatment plans, insurance claims, patient follow-ups between therapy sessions, and new client acquisition. In supporting therapists, Klarify helps optimize for therapeutic outcomes.

• Klarify just did an official launch with Y Combinator — you can learn more here. If you know any therapists who might benefit from using Klarify, or if you’re interested in investing in Klarify’s upcoming round, please reply to this email and I’ll connect you with Moody, the founder and CEO.

More of my work 👇🏼

🎤 Decode videos | For top Berkeley and Stanford founders

• DECODE is the largest founder community co-hosted across UC Berkeley and Stanford. The DECODE annual conference focuses on helping founders in their earliest stages of starting a startup.

• I’ve had the honor of serving on the DECODE board since 2016. You can find videos from the latest conference I helped host here 👇🏼

🤖 Founders' prompt library

• My team and I have crowdsourced the largest AI prompt library optimized for founders – especially fractional founders — to help them build, sell and operate 10x faster and better.

👥 Pop up board

• I regularly host pop-up boards, a unique 45-minute session where founders ask their toughest strategic questions and get tailored advice from top builders and operators from Google, Microsoft, Meta and more.