David Crawshaw

Co-Founder & CEO, exe.dev; Co-Founder, Tailscale

Reviewed Updated Aug 3, 2026

This profile is AI-generated. If you spot an error, please help us fix it by sharing a URL to the correct information.

Systems engineer, co-founder and former CTO of Tailscale, and co-founder/CEO of exe.dev (raised $25M Series A from Amplify, CRV, and Heavybit in 2026). A long-time Go language contributor from Google who writes influentially on LLM-assisted programming, SQLite, and cloud infrastructure. No public angel portfolio; profiled as a technical node in the developer-tools/infrastructure graph rather than as an active check-writer.

Location Berkeley, California
Last Verified Investment exe.dev (Co-founder) — ~2026

Background

David Crawshaw is a systems engineer, co-founder of Tailscale, and co-founder & CEO of exe.dev. He is based in Berkeley, California 1. He previously worked at Google as a Staff Software Engineer, where he specialized in petabyte-scale logs processing, implemented TCP/IP networking for Fuchsia, and led the port of the Go programming language to iOS and Android 12. He was a member of the Go language contributor team, with commits to golang.org, and presented “Go Build Modes” at GopherCon 2017 2. He left Google in March 2018 3.

In 2019 Crawshaw co-founded Tailscale — the WireGuard-based mesh networking company headquartered in Toronto — alongside Avery Pennarun, David Carney, and Brad Fitzpatrick 4. He served as Tailscale’s founding CTO. In a February 2025 interview on the Changelog podcast, Crawshaw stated: “I stepped back from the CTO last year” and described spending his time “exploring sort of new product spaces, things that can be done, both inside and outside of Tailscale” 5.

That exploration became exe.dev, a developer-and-agent-focused cloud infrastructure company where Crawshaw is co-founder and CEO. In April 2026, exe.dev announced a $25M Series A led by Amplify Partners with participation from CRV and Heavybit 67. exe.dev’s earlier internal work was described in Crawshaw’s April 2026 post “Building a cloud” 7.

Crawshaw writes regularly at crawshaw.io on topics including Go, SQLite, compiler optimization, agent-assisted programming, and cloud infrastructure design 3. His 2025 blog post “How I program with LLMs” was widely circulated in developer communities 8.

On investing activity: Crawshaw is primarily an engineer, writer, and operating founder. No independently sourced public angel investments were found in this research pass. He is not listed in the swyx devtools-angels community registry 9. He is profiled here because his writing, technical influence in the Go / developer-tools ecosystem, and role as a repeat founder make him a meaningful node in the investor-adjacent graph — and because founders raising in developer-tools/infrastructure often seek him out as an advisor. If he has done private angel checks, they are not publicly reported.

Stated Thesis

Crawshaw has not published a formal investing thesis. As an operator, his stated product thesis at exe.dev is that AI agents and developers need the same primitive — a computer — and that current cloud abstractions get in the way. He has articulated this repeatedly:

  • On why exe.dev exists: “Agents are trained on how developers work. They want exactly what we want. Full computers, understandable and stable building blocks, familiar systems wherever possible.” 6
  • On cloud VMs: “VMs are the wrong shape because they are tied to CPU/memory resources. I want to buy some CPUs, memory, and disk, and run VMs on it.” 7
  • On the current cloud pricing model: “The standard price for a GB of egress from a cloud provider is 10x what you pay racking a server in a normal data center.” 7

At Tailscale, his stated technical thesis was that mesh networking and end-to-end encryption should replace perimeter security: “Both of these problems have much more elegant solutions if we can trust the network” 10 and Tailscale is “by design, we never see your unencrypted traffic” 10.

Inferred Thesis

No verifiable public angel portfolio was found in this research pass, so a data-grounded inferred thesis cannot be constructed. Qualitative signals from Crawshaw’s public writing and product work suggest he is most likely to be sympathetic to:

  • Developer-first infrastructure: cloud, compute, storage, networking primitives designed for programmers rather than platform operators 67.
  • AI/agent tooling with a systems foundation: he has described “an enormous amount of traditional engineering to do in front of LLMs” 5 and has published extensively on programming with agents 8.
  • Go and SQLite ecosystems: he is a long-standing Go contributor 2 and has open-sourced SQLite tooling on GitHub 11.
  • Zero-trust networking and security-by-default: consistent with his Tailscale co-founding thesis 10.

This is a qualitative inference from writing and product choices, not from counted portfolio data. Founders should not treat this as a proxy for check-writing behavior.

Portfolio

No independently verified angel investments were identified in this research pass. Crawshaw is a co-founder and operator at Tailscale and exe.dev; both are companies he built, not companies he backed as an outside investor.

Company Year Stage Role Source
Tailscale 2019 Co-founder Co-Founder, ex-CTO 4
exe.dev 2026 Co-founder Co-Founder, CEO 67

If you have documentation of angel investments by David Crawshaw (SEC filings, cap-table disclosures, or press coverage naming him personally as an investor), please submit a source via the pending-sources workflow.

In Their Own Words

On agents needing computers (exe.dev Series A announcement):

“Agents are trained on how developers work. They want exactly what we want. Full computers, understandable and stable building blocks, familiar systems wherever possible.” 6

On the shape of cloud VMs:

“VMs are the wrong shape because they are tied to CPU/memory resources. I want to buy some CPUs, memory, and disk, and run VMs on it.” 7

On egress pricing:

“The standard price for a GB of egress from a cloud provider is 10x what you pay racking a server in a normal data center.” 7

On stepping back from Tailscale CTO:

“I stepped back from the CTO last year.” 5

“I am spending my time exploring sort of new product spaces, things that can be done, both inside and outside of Tailscale.” 5

On the current state of AI-assisted coding (Feb 2026):

“In February this year, the latest Opus model can write nine tenths of my code.” 8

On the demise of the IDE in his workflow:

“In 2021, the IDE had won. In 2026, I don’t use an IDE any more.” 8

“The only IDE-like feature I use today is go-to-def, which neovim is capable of with little configuration.” 8

On model selection:

“If you try some penny-saving cheap model like Sonnet, or a second rate local model, you do worse than waste your time.” 8

“You will not know what models will be capable of unless you use the best.” 8

On engineering LLM products:

“There has to be a way to make this easier. And my main conclusion from all of that is there’s an enormous amount of traditional engineering to do in front of LLMs to get there.” 5

“They are actually extremely hard to turn into products, and to get those details right, in a general sense, for shipping to users.” 5

On chat interfaces:

“Chat is currently our primary user interface on the models, and it’s not the best interface for most things.” 5

On the Tailscale founding thesis (2021 interview):

“Both of these problems have much more elegant solutions if we can trust the network.” 10

“by design, we never see your unencrypted traffic” 10

On startups and language choice (ELC podcast):

“If you’re a pre-seed startup in the Bay Area and you are trying to ship something to your first few customers and you ship with 40-something languages, then you’ve made a mistake, but it’s not necessarily true that a large company has made that mistake.” 12

On his craft (Uses This interview):

“I like to write code and help others write code.” 13

What Founders Say

No independently sourced founder testimonials found. Crawshaw does not appear to have a publicly documented angel portfolio, and no third-party founder quotes describing him as an investor or advisor were surfaced in this research pass.

Sources


  1. Crunchbase, “David Crawshaw – CTO and Co-Founder @ Tailscale,” accessed August 2026. https://www.crunchbase.com/person/david-crawshaw

  2. Google Go language contributors list, accessed August 2026. https://go.dev/CONTRIBUTORS

  3. crawshaw.io blog index, accessed August 2026. https://crawshaw.io/blog/

  4. Wikipedia, “Tailscale,” accessed August 2026. https://en.wikipedia.org/wiki/Tailscale

  5. Changelog Interviews #629, “Programming with LLMs featuring David Crawshaw,” accessed August 2026. https://changelog.com/podcast/629

  6. exe.dev blog, “Series A for exe.dev,” April 22, 2026, accessed August 2026. https://blog.exe.dev/series-a

  7. David Crawshaw, “Building a cloud,” crawshaw.io, April 22, 2026, accessed August 2026. https://crawshaw.io/blog/building-a-cloud

  8. David Crawshaw, “Eight more months of agents,” crawshaw.io, February 8, 2026, accessed August 2026. https://crawshaw.io/blog/eight-more-months-of-agents

  9. swyx, “devtools-angels” GitHub registry, accessed August 2026 (Crawshaw not listed). https://github.com/swyxio/devtools-angels

  10. Tailscale blog, “We all have to do a better job managing our infrastructure” (Cybernews interview with David Crawshaw), accessed August 2026. https://tailscale.com/blog/cybernews-interview

  11. David Crawshaw GitHub profile, accessed August 2026. https://github.com/crawshaw

  12. Engineering Leadership Community podcast, “Rapid prototyping & developing your product instinct” with David Crawshaw, accessed August 2026. https://sfelc.com/podcasts/rapid-prototyping-and-developing-your-product-instinct-david-crawshaw-tailscale

  13. Uses This, “David Crawshaw” interview, accessed August 2026. https://usesthis.com/interviews/david.crawshaw/