Vaughn Vernon
banner
vaughnvernon.bsky.social
Vaughn Vernon
@vaughnvernon.bsky.social
Software Architect and Modeler | #DDDesign | Systems Transformation | Simplicity | Writes Code | Actor Model | @kalele_io @kalele_domo | em dashes—my own
Q on LinkedIn: How do you achieve fault-tolerance in a runtime like NodeJS that wasn't designed for it?

A: #DomoActors-TS is not designed for Node.js. It's designed for any JavaScript runtime, the primary one being V8 (@cloudflare.social Workers and Durable Objects), ...

1/3
I have published #DomoActors for TypeScript, an #ActorModel toolkit for the JavaScript platform with fault-tolerant, message-driven concurrency.

$ npm -install domo-actors

See below for access. 1/3
November 14, 2025 at 3:39 PM
Reposted by Vaughn Vernon
Very cool, we'll put it on our library watch segment!
November 14, 2025 at 4:21 AM
I have published #DomoActors for TypeScript, an #ActorModel toolkit for the JavaScript platform with fault-tolerant, message-driven concurrency.

$ npm -install domo-actors

See below for access. 1/3
November 14, 2025 at 12:31 AM
When I read the headline, "Does OpenAI Need a Bailout?"
a man in a plaid shirt and tie is making a surprised face .
ALT: a man in a plaid shirt and tie is making a surprised face .
media.tenor.com
November 10, 2025 at 6:50 PM
I decided to change the SchemaMinder open source license to the Reciprocal Public License 1.5 (RPL 1.5).

See links to repository and license text.

github.com/VaughnVernon...

opensource.org/license/rpl-...
November 7, 2025 at 3:04 PM
I decided to change the SchemaMinder open source license to the Reciprocal Public License 1.5 (RPL 1.5).

See:

github.com/VaughnVernon...

opensource.org/license/rpl-...
GitHub - VaughnVernon/SchemaMinder: A multi-tenant schema registry.
A multi-tenant schema registry. Contribute to VaughnVernon/SchemaMinder development by creating an account on GitHub.
github.com
November 7, 2025 at 2:48 PM
If there's anything that has become clear to me while using #GenAI, it's this:

AI speed is abundant. Craft is scarce.

Those inexperienced in software development while embracing GenAI may never understand how software is best constructed. 1/3
November 4, 2025 at 2:17 AM
There are many new topics and tools to share in my upcoming "Implementing Domain-Driven Design Workshop"

It starts in just one month. Register today.

kalele.io/iddd-workshop
Register for my upcoming "Implementing Domain-Driven Design Workshop" December 8-11 and learn the latest on #DDDesign strategic and tactical modeling with #AI #LLM assistance on #Serverless. Yes, we're going there together.

It's not just time. It's deeper insights faster.

kalele.io/iddd-workshop
November 3, 2025 at 3:35 PM
Reposted by Vaughn Vernon
Register for my upcoming "Implementing Domain-Driven Design Workshop" December 8-11 and learn the latest on #DDDesign strategic and tactical modeling with #AI #LLM assistance on #Serverless. Yes, we're going there together.

It's not just time. It's deeper insights faster.

kalele.io/iddd-workshop
October 30, 2025 at 3:44 PM
Let's play Jeopardy!

A: OpenAI's IPO
.
.
.
.
.
.
.
.
.
.
Q: What is Ocean's 13 x 1,333.333333333
October 31, 2025 at 4:58 AM
Register for my upcoming "Implementing Domain-Driven Design Workshop" December 8-11 and learn the latest on #DDDesign strategic and tactical modeling with #AI #LLM assistance on #Serverless. Yes, we're going there together.

It's not just time. It's deeper insights faster.

kalele.io/iddd-workshop
October 30, 2025 at 3:44 PM
I adore this reply:

"These answers thoroughly address all my questions and concerns. The additional context changes the entire design calculus" 😍
October 24, 2025 at 11:54 PM
This is an amazing prediction by #CarlHewitt, inventor of the #ActorModel, regarding a revolution in #AI by 2025. It's sad that like many of the early pioneers, he didn't live to see it happen. (Video: 5 minutes.)

www.youtube.com/watch?v=A69u...
Let’s #TalkConcurrency with Carl Hewitt
YouTube video by Erlang Solutions
www.youtube.com
October 23, 2025 at 4:11 PM
I have improved the set up of the #SchemaMinder project that demonstrates the results of using @anthropic.com #ClaudeCode and @cloudflare.social serverless.

See: docs/GETTING_STARTED.md

Use: npm run setup

That gets you started quickly.
October 22, 2025 at 6:52 PM
The is my #OSS #SchemaMinder serving as my public-facing exploration and experiment with @anthropic.com #ClaudeCode. It runs on @cloudflare.social and a work-in-progress, but definitely usable. See the brief README with references to other resources.

github.com/VaughnVernon...
GitHub - VaughnVernon/SchemaMinder: A multi-tenant schema registry.
A multi-tenant schema registry. Contribute to VaughnVernon/SchemaMinder development by creating an account on GitHub.
github.com
October 21, 2025 at 6:28 PM
Smalltalk went in a direction that Kay was not happy with. It seems to have become something like design-by-committee. He stated that somewhere; whether conference talk, interview, or written, I don't recall.

He seems to have influenced others, Carl Hewitt being one.

1/3
Some are wondering why I said that "true OOP was lost from C++ onward." I don't have time or space to go in-depth, but here are a few Alan Kay quotes: 1/5
There are a number of useful paradigms. OOP and FP both have merit. Real OOP was lost from C++ onward.

Consider: "Functional Core, Imperative Shell"

"Strategic Monoliths and Microservices: Driving Innovation Using Purposeful Architecture" See Ch8 & 9.

www.informit.com/store/strate...
October 20, 2025 at 6:43 PM
Some are wondering why I said that "true OOP was lost from C++ onward." I don't have time or space to go in-depth, but here are a few Alan Kay quotes: 1/5
October 20, 2025 at 4:47 AM
There are a number of useful paradigms. OOP and FP both have merit. Real OOP was lost from C++ onward.

Consider: "Functional Core, Imperative Shell"

"Strategic Monoliths and Microservices: Driving Innovation Using Purposeful Architecture" See Ch8 & 9.

www.informit.com/store/strate...
Strategic Monoliths and Microservices: Driving Innovation Using Purposeful Architecture | InformIT
Make Architecture Choices That Free You to Maximize Value and Innovation"The heart of this book is a large set of thinking tools that will help you design a new architecture . . . and the organization...
www.informit.com
October 19, 2025 at 11:42 PM
After a month away on other work, I finally got back to my product and have it in fairly good shape. It's generated by Claude Code under my architecture and design guidance. Here's a screenshot of the UI. I am not going to polish it much more, so I hope to open the repo sometime during the weekend.
October 18, 2025 at 5:50 AM
"8 SaaS Niches Printing $20K-Plus a Month — And Nobody’s Built Them Yet"

You can't make this stuff up unless you wrote the post, in which case you did make this stuff up.
October 14, 2025 at 4:37 AM
Aren't Sprints bad?

Maybe. I think that defining a Sprint as a single iteration is silly. It makes an iteration very mechanical, as if there's no progress made until the timebox officially ends.

In reality, even in a single-week Sprint timebox, there are likely several iterations. 1/5
There is no single part of Scrum that is bad. It's the inseparableness of and strict adherence to all the parts of Scrum that is bad. By definition, you can't separate any parts of Scrum, as in use a subset of Scrum linked with common sense, and still call it Scrum. 1/6

Figure: Not complex enough.
October 13, 2025 at 7:22 PM
There is no single part of Scrum that is bad. It's the inseparableness of and strict adherence to all the parts of Scrum that is bad. By definition, you can't separate any parts of Scrum, as in use a subset of Scrum linked with common sense, and still call it Scrum. 1/6

Figure: Not complex enough.
October 13, 2025 at 7:07 PM
I'd like to make something perfectly clear. This kid covers Tool's full discography in a single take—7.5 hours, nonstop.

Thank you. That'll be all.

youtu.be/0kO0SnAu8Nc
TOOL's Discography in One Take (Drum Cover) - Tyler Visser
YouTube video by Tyler Visser
youtu.be
October 12, 2025 at 11:21 PM
Being LLMs that run on GPUs that can't perform associative math, there's no reason to expect a given prediction. We want consistency, but the LLM platforms are inconsistency machines by design.
October 8, 2025 at 7:11 PM
Our book "Strategic Monoliths and Microservices" doesn't categorically (as an absolute) recommend modular monoliths. The key is in the subtitle: "Driving Innovation Using Purposeful Architecture." 1/4
October 8, 2025 at 6:58 PM