The Robots are coming for your job, and that's okay - Elmer Thomas and Maria Bermudez

Hello everyone, I'm Elmer Thomas, Principal Developer Educator at Twilio. Welcome to the AI Engineer World's Fair. Our talk today is called "The Robots Are Coming for Your Job, and That's Okay." Don't worry, the title is tongue-in-cheek. I'm here to show you why AI isn't something to fear; it's something we can embrace to work smarter.

In the next few minutes you'll see how our relatively tiny docs team uses AI agents to turbocharge workflows instead of trying to replace ourselves. We'll look at the pain points, the agents we built, and the guardrails that keep us accurate.

I've been a developer and builder for decades; trends come and go. AI is the latest big one. Like many of you, I wondered how it would change our day-to-day work. Today you'll see what we've learned, warts and all.

My colleague Maria Bermudez, a lead developer of our AI docs buddy and resident backlog-busting coffee fanatic, will drive a live agent walkthrough so you can judge the results yourself. Here's a game plan: we'll cover the pain, our agent architecture, Maria's demo, and a roadmap that you can steal for your own team.

Let's dive in. Picture this: a tiny docs team, a flood of Jira tickets. Three headaches jump out. Number 1: error-prone first drafts from 100+ product teams. Two: time sync grooming, like style checks, alt text, SEO. Three: hallucination risk if we let Gen I run wild.

We needed leverage, not burnout. Instead of one mega-bot, we built 6 single-purpose agents behind a simple Next.js front-end. The automated editor fixes grammar, formatting, and accuracy. The image alt text generator gives you instant accessibility wins. The jargon simplifier turns DevSpeak into plain English.

SEO metadata provides title and description, and keeping it character-count safe. The docs outline builder recommends nav and structure. That's coming soon; we're super excited about that. Our Slack backbot helps triage help channel requests. Each agent tackles a repetitive, well-scoped job so humans can focus on judgment and clarity.

Rule of thumb: pick tasks that are repeatable, high volume, and low creativity. That's the sweet spot for an AI helper. Here's the flow behind every request: we have a Next.js UI that feeds into a custom GPT-4/01 agent. We use the appropriate model for the particular job.

The custom GPT has our style guide, and it has our rubric that is baked in. It's actually retrieved from an Airtable so that folks can easily collaborate on it. We have a validation layer that includes veil linting and CI/CD tests.

The GitHub PR adds code owner review and makes it much easier for us to scrutinize the changes that the agents are suggesting. A human hits the merge only when it'sright, and usually we have built into their product and engineering reviews as well.

So we have several human eyes before we actually consider hitting that merge button. This layered approach, it crushes hallucinations without showing us down. Maybe not crushes them, but it reduces them significantly. Alright, it's time to prove this isn't vaporware.

I'm handing the screen to Maria Bermudez, the brains behind much of the AI docs buddy code. Maria, take it away.

Thank you, Elmer. So, real quick, I want to show you how we use Copilot to auto-generate our overview and release notes page. And from here you can get into any of the agents that you want to work with.

So I'm going to start with the automated editor. One cool thing about this is you not only can load an MDX file, or Markdown file, but you can also plug in a live URL. And I'm going to do that optionright now.

This does take a little while, so we added this graphic here because we're using the 01 model. We've done a lot of experimentation and figured out this is the best model to consistently apply our style guide and rubric.

And here you go. And then on the

it shows you the diff of what it changed. And in the changes made tab, you actually have a list of explanation of the original text, the revised text, and what it changed, and what style guidance and rubric item the changes pertain to.

It's not perfect, as you can see. Like, sometimes it catches that there's a missing SEO description, but other times it doesn't. So luckily we have an SEO metadata generator. And I'm going to plug in this page into that one.

So what I'm actually going to do is I'm going to upload a file, and then go back to conversational intelligence.

So now I'm going to generate SEO metadata for that same page that we just edited.

And there you go. So it generates a meta title, so you have to specify if you don't want that. And then it gives you the meta description, accounting for character limitations. And now I'm going to show you the alt text generator.

It has the same option that we have where you can plug in a live URL, but I'm going to show you picking multiple pages and

having our agent generate alt text for all of those images. So I'm going to click on generate alt text now.

And you see how quickly that got done. And it also conforms to the format that we require from our docs platform. And lastly, I'm going to show you the jargon simplifier. So you can take whatever, like, pages that you're reviewing.

This is really helpful not only for writing, but when we're reviewing a pull request. And it gives us, like, a good way to provide quick edits to those images. So I'm going to plug in a text that I've prepared, and I'm going to click on simplify.

And there you go. It also gives you a diff, like our editor. And on the next tab you have the revised text that you can quickly copy and apply as a pull request comment, or just edit the file directly.

And that's it. We're working on making our agents talk to each other eventually, but we have a lot of work to do on that front.

Thank you, Maria. That was wonderful. Now, tools alone don't guarantee quality. Our guardrails tackle three big risks: hallucinations. We use tools like veil lint and CI test to help mitigate that, in combination with the various eyes of human stakeholders.

Bias: dataset tests and prompt audits are the way to tackle those issues. Stakeholder misalignment: so weekly PR reviews, sometimes these are compressed into days and sometimes hours. And we also have Slack feedback loops, generally with the product managers and engineering team.

These feedback cycles, they let us tune the prompts continuously instead of hoping the model magically stays perfect.

Now, please steal this three-step playbook. Identify one pain that's killing throughput. Two: pick a single task that's repeatable and rule-based. And then three: loop with your users weekly at least. Ship, measure, and refine. Stack a few of these wins and watch your team's velocity jump.

Thanks again to the AI Engineer World's Fair team for hosting us. And thanks to you for spending time with us today. Big shout-out to Carly VanderWert, our manager at Twilio, who keeps encouraging us to push the boundaries with AI, along with the rest of our leadership.

We appreciate your attention. Now go build something amazing with AI.