The AI Record
We built an architecture.md file. Instead of using … MD, just pick something that every model can just understand. This file has to be incredibly small, and it can only have things that will not change for months or for years. In our case, it's the layers of the compiler. … So now we have standards so anyone can use whatever they want.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing how the team sets shared standards when engineers choose their own AI tools.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 2:10.

Checked against the source [ Grade B ] Said in the last year Sells AI tools

Filed under

Who is speaking

Vaibhav Gupta, Co-founder and Chief Executive. Runs the company and describes its own engineering practices and the language it builds; Boundary sells AI tooling and has reason to present its approach favorably.

Boundary (BoundaryML) sells AI or agent tooling, so it has a reason to publish flattering numbers about AI. The passage stays in the archive; read it with that in mind.

How this was checked

Checked against the source by a first reader, then confirmed independently by a second reader who fetched the source without seeing the first reader's notes (Independent review agent (Claude Sonnet), blind second pass from passage text and citation only, no repository access, 2026-11-01).

Whisper SRT 00:02:10-00:02:33. Whisper has 'Cloud MD' and the passage has '… MD', so the '…' covers the omitted word. The omitted compiler sentences are also covered. Matches.

How to cite this passage

Vaibhav Gupta (Boundary (BoundaryML)), “Fighting Slop with Slop”, AI Engineer (YouTube), 2026-07-31, at 2:10. https://www.youtube.com/watch?v=AMiyLItEtLA Archived at https://the-ai-record.com/passages/boundary-architecture-md-small-stable-layers-compiler-2026

More from the same source

You could figure out which ones took less tool calls, … which one made less errors, which one produced the correct outcome, and deterministically know what's going on. The point is you can start building data-driven systems without ever writing a single line of code. … in order to build a programming language, it wouldn't have taken eight people. It wouldn't have taken less than two years. It would have taken hundreds and thousands and tens of thousands of man hours, and then you would still have a broken system. And today, we can just spend billions of tokens and make it work, and we can make it stable.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is comparing what building a programming language would have taken before with what it takes the team now.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 6:00.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

What we did was we built a system that actually has agents constantly running and creating BAML programs. … We then look at the entire … transcripts, see what tools they use, see what happened, and obviously we as humans can inspect them. But more importantly, we can have agents go inspect them. And agents find what was good, what was bad. … And we can have humans collaborate with these issues to figure out which ones are real, which ones are hallucinations, which ones don't have taste as much as I hate to use that word. And then we can have agents go ahead and create fixes to these problems and go to address them.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing how the team tests its programming language when it does not rely on reading every line of code.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 5:05.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

But the real problem with all of this is I built this and I hit a little bit of AI psychosis. And I started shipping 10 design docs a day. And soon, the team was fighting my slop. … This last rule was … ship a design doc, you require people to actually go read it. And with this last standard, we suddenly had design docs … incredibly high quality.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing the design-document tool he built for the team and the rule that followed.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 3:37.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

This tool basically visualizes our dependency graph internally with some external dependencies as well, … But what's more interesting is we can go build CLI tools that guarantee that certain invariants can't be broken. And what this does is when Claude builds a new package or adds a dependency that's leaky, we now have CI/CD changing or a simple git commit history that tells us exactly where things break. And by this, we're actually able to make our architecture change. We haven't changed our architecture in the last three or four months.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing how the team keeps the BAML code base from drifting when engineers do not review each other's code.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 4:05.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

we have a very simple rule in our team. Code can be slop, writing cannot. … What this design doc tool does, it's a replacement for both Notion and GitHub, effectively for design docs. It allows versioning, commenting, all the stuff you want. … We built another tool on top of that. And this tool was a slack integration for that tool. Every time a design doc got updated, this channel got notifications. And what ended up happening is this channel became the most popular channel in our company really fast.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing how the team keeps design documents good while letting code be generated.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 2:43.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

When we started building BAML, I didn't think it would be possible to build some of the software we did. And just yesterday, one of … engineers built a partial C compiler, purely in BAML. … Our processes have to evolve if we're going to ship at agent speed.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Gupta is describing what the team has been able to build in BAML, the programming language it makes.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 19:47.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

We do no code reviews. We require every engineer to work on things in parallel, and we have no standardization how people do AI.

Vaibhav GuptaCo-founder and Chief Executive · Executive, Boundary (BoundaryML)

Context. Opening of a conference talk by the chief executive of Boundary, the company that builds the BAML programming language, on how its team works.

Talk: Fighting Slop with Slop, AI Engineer (YouTube), July 31, 2026, at 0:26.

Checked against the source [ Grade B ] Said in the last year Sells AI tools Passage page

Something wrong? Report a correction. JSON: /api/v1/passages/boundary-architecture-md-small-stable-layers-compiler-2026.json