How to get your career website into AI: Shaker’s honest answer

By Jeremy Breit, Director of Web Development & Solutions at Shaker Recruitment Marketing. Jeremy has 10 years of experience in the career website industry and leads Shaker’s career website offering for some of the world’s leading employers.

Almost every client I talk to asks me some version of the same question: “How do we show up in AI?” It’s the hottest topic in recruitment marketing right now, and I get it: candidates are typing “what’s it like to work at [Company]?” into ChatGPT and getting an answer before they ever land on your career site. That’s a little terrifying.

I’m going to give you my honest take. And if you’re wondering whether AI helped me write this — here’s an em dash —

I’ll also be updating this every couple of months, because this space is moving fast and I’d rather tell you “this changed” than let outdated advice sit on the internet forever.

First, let’s be honest: there is no magic button

Before I give you a checklist, I want to set expectations. There is no single, definitive thing you can do today to guarantee you show up in every AI tool. Anyone who tells you otherwise is probably also selling you something.

What most people are calling “showing up in AI” is really a mix of two things the industry is starting to call Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO). In simple terms: AEO is about whether your content can be surfaced as a direct answer to a candidate’s question. GEO is about how AI models interpret and represent your brand overall. They overlap, and most of what I’m going to cover touches both.

Here’s why there’s no single fix: not all “AI” works the same way.

Trained LLMs (ChatGPT, Claude, Gemini) learned from massive snapshots of the web taken months or years ago. What they know about your company was baked in during training. You can’t update that retroactively, and no amount of schema markup will pipe your latest content directly into GPT-4’s brain.

LLM-powered search tools (Perplexity, ChatGPT with Search, Google AI Overviews) are a different animal. These crawl the web in near real-time and use your current content to generate answers. This is where your optimization work pays off most directly and most immediately.

Knowing which type of AI you’re trying to influence matters. Most of what I’m going to recommend is relevant to both, but I want you to understand why it works, not just what to do.

Your career site still matters. Maybe more than ever.

I’m biased, obviously. I build career sites for a living. But here’s my genuine argument for why your career site is more important in the age of AI, not less.

LLMs are operating on snapshots. They are, by nature, behind. A candidate who gets a response from ChatGPT today might be reading information that’s six months, a year, or two years old. Wrong office locations. Outdated benefits. A culture description from before a major reorg.

Your career site is your source of truth. It’s the one place you control completely, that gets crawled regularly, and that a candidate can land on and say “okay, this is current, this is real.” As AI-generated answers become more common, candidates are going to develop habits of double-checking, and your career site needs to be ready to be that final confirmation.

Think of it this way: AI gets them curious. Your career site closes the deal.

The technical foundation still applies

I work on career sites across a lot of platforms, custom builds and partner platforms alike. The technical optimization principles that have always mattered for SEO apply directly to LLM-powered search tools as well.

Advanced JSON-LD schema on every page. Structured data helps crawlers understand your content quickly and accurately. Organization schema, FAQ schema, and proper breadcrumb schema on your career site give AI-powered tools clean, machine-readable signals about who you are and what you offer.

Google for Jobs schema on every job description page. This one is non-negotiable. If your jobs aren’t tagged with proper JobPosting schema, you’re invisible in both traditional search and AI-powered job discovery.

Internal linking. A well-linked career site lets crawlers (AI or otherwise) discover depth. Your life in [City] page should link to your benefits page, your team pages, your relevant job categories. Don’t make crawlers work for it.

A real caveat here: if your career site lives on a partner platform, you may have limited control over some of these. That’s a reality of the ecosystem. Platforms vary significantly in how much they expose to you. What I can do is make sure we push as far as the platform allows and document what’s constrained so you know exactly where you stand.

One thing many teams don’t think about: your robots.txt file. AI crawlers like GPTBot and ClaudeBot respect robots.txt exclusions. If your site is inadvertently blocking them (which some platforms do by default) you’re opting yourself out of the conversation entirely. It’s worth checking intentionally, not accidentally.

Content is where most career sites actually lose

Technical stuff is table stakes. Content is where the real differentiation happens, and where most career sites are genuinely failing.

Tell a story that brings your EVP to life. I can’t tell you how many times I’ve seen a career site where the content is either copy-pasted from the corporate site or lifted word-for-word from a five-year-old site refresh. LLMs are looking for authentic, specific, textured content. “We’re a collaborative, innovative team” tells a model nothing. “Our engineering teams ship to production on day one and run their own on-call rotations” is something a model can actually use.

Your headers need to do real work. Structure matters, and that means thinking about your H1, H2s, and H3s as content, not decoration. Not just “Our Culture” or “Benefits,” but “What Flexible Work Actually Looks Like Here” or “How We Support Career Growth at Every Level.” Descriptive, tactical headers help humans scan and give AI crawlers meaningful signals about what your content covers. Look at the section headings in this blog post. Each one tells you exactly what you’re about to read. Your career site pages should work the same way.

Job description pages need relevant content beyond the job itself. If a candidate lands on a Software Engineer JD, they shouldn’t just see the job. They should see something about your engineering culture, links to related roles, and a window into what the team looks like. That context matters both to the candidate and to the crawlers indexing the page.

Your job descriptions are hundreds of pages of dynamic content

Here’s something career site owners undersell: if your jobs live on your career site, you have a constantly updating library of pages. Every job posted is a new, indexable URL with fresh content. That’s a natural SEO and crawlability engine that hosted ATS solutions can’t replicate.

But only if the job descriptions are actually good.

Outdated postings, vague descriptions, walls of requirements with no context. These don’t just frustrate candidates, they give crawlers nothing to work with. A well-formatted, descriptive job description with a clear title, structured responsibilities, realistic qualifications, and some texture about the team and role is the kind of content that performs. Keep job descriptions current. Format them properly. Don’t let them turn into zombie postings.

What the rest of the internet says about you matters too

Your career site is important. But LLMs don’t just read your career site. Glassdoor, Indeed, LinkedIn, Reddit, old press coverage: all of it feeds the narrative. If those sources are outdated, off-message, or negative, that’s what AI is working with regardless of how good your career site is.

The clearer and more uniform your messaging is across your career site, job ads, social, and third-party platforms, the stronger your overall signal. AI rewards consistency. Siloed employer branding is a liability in an AI world, and most companies have no idea where the gaps actually are. Which is exactly what the next section is about.

Know what AI is saying about you

Everything above is about improving what AI finds when it looks. But before you can fix any of it, you need to know what AI is actually saying about you right now. Most companies don’t. They assume because their career site looks good or their Glassdoor rating is decent that the story AI is telling about them is positive, accurate, and competitive. Often it isn’t.

AI tools are forming opinions about your employer brand every time a candidate asks a question. What it’s like to work there. How you compare to a competitor. Whether your culture is real or just marketing copy. Those answers are being assembled from dozens of sources across the web, and the signals those sources send, consistent or contradictory, current or outdated, specific or vague, determine what candidates hear before they ever talk to a recruiter.

Understanding your AI presence means understanding those signals: where they’re coming from, what narrative they’re adding up to, and where the gaps are between what you want candidates to know and what AI is actually telling them. This is an area Shaker is actively researching and developing every day, and it’s becoming a foundational part of how we think about career site strategy going forward.

A note on partner solutions

Some of our partners have also built products specifically designed to get your open roles in front of AI-powered job discovery tools. If you’re looking to improve how your jobs surface when candidates are actively searching, those solutions are worth a conversation too.

If you’re curious about whether our internal or partnership solutions apply to your setup, ask me.

The short version

You can’t guarantee AI knows you. But you can make sure every surface where AI looks is telling a consistent, credible, current story, and that your career site is there when a candidate wants to verify it for themselves.

I’ll update this as the landscape shifts. Disagree with anything here? Shoot me an email and let’s chat!