What happens when you give the smartest people an unlimited budget to solve impossible problems?
Google might be the closest answer we have to that hypothetical.
The tech giant today has over 180,000 employees, around half of them in Engineering. But what’s striking to me is how the most impactful technical breakthroughs, especially in their “scale-up” phase from 2000 to 2015, often came from just one or a few select individuals: The 5-person Gmail team, the PM rewriting Google Maps over the weekend, 4 days of pair programming to optimise the entire index, etc.
You could argue that nearly half of Google’s value today (~$2 trillion) can be directly attributed to the top 9 engineers throughout its history.
They are the Google Fellows:
Urs Hölzle architected Google’s internal cloud, turning servers from a cost centre into a significant competitive advantage. His early work on file systems, cluster scheduling, and networking carried over to their external cloud (GCP, today a ~$50B annual run rate business on its own).
If your team mandates peer code reviews or writes postmortems, you can thank Hölzle too. He introduced these practices at Google as employee #8. The engineering culture he set would be copied by a generation of Silicon Valley startups.
Luiz André Barroso practically invented the modern data centre as we know it. He co-authored “The Datacenter as a Computer” with Hölzle, which pushed for co-designing hardware, software, networking for efficiency and performance at scale. Barroso deeply appreciated what he calls the “obvious questions”—In his most-cited paper, he asked: “Shouldn’t servers use little energy when they’re doing little work?”
Amit Singhal became obsessed with informational retrieval after studying it under “the father of digital search” at Cornell. He rewrote the core search algorithm (expanding on the OG PageRank) in 2001 then led Google Search for the next 15 years.
R. V. Guha co-created the Resource Description Framework (RDF) and Really Simple Syndication (RSS), which set the standard for how data moves on the web. His work on Schema.org further organised the world’s information and helped Google index & serve more relevant results.
Ramakrishnan Srikant was one of the first to apply ML to improve ad systems (quality, targeting, prediction) and built their early internal ML infra. Search accounts for 58% of Alphabet’s total revenue (~$200B annual run rate), so even the tiniest optimisations that get the right ads to appear end up being insanely valuable.
Sebastian Thrun is a renowned researcher in AI, robotics, and autonomous systems. While at Stanford, he led the team that won the 2005 DARPA Grand Challenge for driverless cars. He later joined Google to found X, the moonshot factory that spawned Waymo (most recently valued at $45B
