5 ways career advisors are using AI every day, and what it says about the system around them
Five workflows career advisors built themselves. And the one they couldn't.
Tolu Towoju
6 min read
Career advisors are quietly building their own AI workflows. Five real examples, and what they reveal about the system around them.
Career advisors are some of the most resourceful people in any workforce program. The industry keeps arguing about whether AI will replace coaching. Meanwhile, working advisors quietly opened a tab, started experimenting, and built workflows that actually save them time.
This piece is about what those workflows look like, in their own words. Five advisors. Five different uses. Read them closely. There's a pattern in them that nobody's named yet, and we'll come back to it.
1. Practicing the hard conversation before it happens
Some conversations need rehearsal. The one where a learner who's been ghosting finally picks up the phone. The one where you have to tell someone their resume is the reason they're not getting callbacks. The one where a learner is convinced they want a role they're not ready for, and you have to disagree with them honestly without losing the relationship.
, HR Consultant and Career Coach (ICF-PCC), , built her own way to prep for these:
Stop guessing who's interview-ready.
See how Clarivue gives career centers the visibility, structure, and proof they need to improve placement outcomes.
I'll set the scene with a chatbot, give it context about the client situation and the kind of challenges they might bring, and then carry on a back-and-forth like an actual coaching session. It doesn't replace the actual human conversation, but it gives me something close enough to practice with. The unexpected turns, the pauses where I have to think on my feet, the chance to try different angles with very low stakes. I do this inside my own private GPT space so that any context I share stays secured.
That last sentence is the part most program directors miss. The privacy concern is real, and Mary solved it herself. Most advisors don't have that option. They either use the tool anyway and hope for the best, or they don't use it at all.
2. Spotting patterns in what learners actually say
Advisors hear the same vague phrases hundreds of times a year. I'm stuck. I just need clarity. I need to figure out what I want. Each one means something different depending on who's saying it. Pattern recognition is most of the job, and it's the part that's hardest to scale.
Kent Vanho runs Alpha Coast and uses AI to do exactly this kind of clustering, just with sales prospects instead of learners. The principle is the same:
Kent Vanho
CEO, Alpha Coast
“
I'll use AI to cluster those phrases into sharper themes, then rewrite messaging so the offer speaks to actual buyer intent instead of generic coaching language. Most coaches I've worked with were relying on referrals or sporadic LinkedIn activity, which creates inconsistent demand. The fastest gains usually don't come from 'more marketing'; they come from clearer positioning that makes the right prospect feel understood the moment they land on your profile.
Swap "buyer" for "learner", and you have one of the most useful AI workflows for an advisor managing a 50-person caseload. I'm stuck from a learner who hasn't started a resume means something very different from I'm stuck from a learner who's done six mock interviews and can't get past round two. Same words, different intervention. AI is good at telling them apart when an advisor doesn't have time to.
3. Turning messy reflection into a clear next step
Most learners don't need more information. They need one specific next move that's small enough they'll actually do it. Advisors know this. But turning a 30-minute conversation into one good action item is harder than it sounds, especially when you're doing it sixty times a week.
Michael Krowne, a renowned entrepreneurship coach, Michael Krowne LLC, who coaches sober founders alongside running four companies, uses AI as a synthesis layer:
Michael Krowne
Founder, Michael Krowne
“
I'll take a client's journal dump, call notes, or a voice memo after a rough week and use ChatGPT to pull out patterns like avoidance, overcommitting, people-pleasing, or decision fatigue. Then I have it convert that into a simple weekly plan: one hard conversation, one tiny habit, one boundary, and one measurable follow-through goal. The key is I don't let AI do the coaching for me. I use it like a fast pattern-recognition layer so I can spend more human energy on judgment, accountability, and helping someone make one real move forward.
His last sentence is the one I'd want every advisor to internalize. AI doesn't coach. It shortens the time between hearing something and knowing what to do with it.
4. Carrying context forward across sessions
This one should make program directors stop reading.
Sarah Barry, a coach and author, describes the workflow that almost every advisor has cobbled together some version of:
Sarah Barry
Founder, Sarah Barry Consulting
“
I use AI to maintain a running document across a coaching relationship, capturing session themes, patterns, and progress over time. Before each session I use it to prep and generate ideas for worksheets or exercises. Afterwards I use it to gather my thoughts into a clear follow-up email for the client. It doesn't replace the coaching. It sharpens the thinking around it.
A running document, manually maintained, on the advisor's own time, in a tool that wasn't built for workforce programs. That is a CRM. It's just a CRM the advisor is building by hand because the actual CRM their program gave them doesn't do this.
If you run a program, ask your advisors what they use to keep track of context between sessions. The answer is going to be some combination of memory, sticky notes, Excel, and whatever AI tool they personally pay for. None of that scales, and none of it shows up in your funder reports.
5. Adapting how you communicate, one learner at a time
Every advisor already knows the same conversation lands differently depending on who's in the chair. The learner who needs the data laid out before they'll commit. The one who needs you to start with the why before the what. The one who shuts down if you push and opens up if you wait. Reading that quickly is part of the craft. Doing it for fifty people a week without losing track of who's who is where it gets hard.
I use Crystal AI to understand a client's communication style and decision-making process before we work together. For someone identified as an analytical thinker, I'll present solutions in a logical, fact-based way. For someone more expressive, I'll lead with energy and creative engagement. It's helped clients feel understood and appreciated, and they're more willing to adopt new strategies when the approach is tailored to them.
The tool isn't the point. The point is that adapting communication style is something good advisors already do instinctively, and AI helps them do it faster. The advisor still has to read the room. The system just shortens the runway.
6. And the one they couldn't build alone: catching the learner who's about to disappear
Every advisor has had this moment. A learner goes quiet between weeks four and seven. They stop replying to texts. They miss two mock sessions in a row. Their resume edits stall after the second round. By the time anyone notices, the cohort has moved on without them.
The frustrating part is that the signs were there the whole time. Missed check-ins. Stalled progress. Response times are getting longer week over week. The pattern is in the data. Most programs just don't have a way to see it before it turns into a missed placement.
This is the workflow advisors can't really build for themselves with a personal AI tool. The signal lives across the program, not inside one advisor's tabs. So advisors do what they always do. They carry the pattern in their head, hope they catch the next one earlier, and feel awful when they don't.
Tolu Towoju
Founder & CEO, Clarivue Intelligence
“
The advisors we work with aren't missing the signs because they're not paying attention. They're missing them because nobody's stitched the signals together. A learner stops replying to one advisor's texts, misses a check-in with another, and skips a workshop a third advisor runs. Each piece looks small on its own. Together it's a pattern. By the time somebody notices, you've usually lost the placement.
This is the workflow that costs the most when it doesn't exist. And it's the one that has to happen at the program layer, not the advisor layer.
The pattern
Look at the six workflows again. Practice. Pattern recognition. Synthesis. Continuity. Adaptation. Risk-spotting.
The first five are advisors building infrastructure for themselves, in a private tab, on their own time, with tools that weren't designed for workforce programs. The sixth is the one they can't build alone, because it has to live at the program layer.
All of them point to the same thing. The argument isn't that advisors need more training, or more AI, or better tools to play with on the side. Advisors are already the smartest people in the room. The question is whether the system around them is keeping up.
The point of an operations layer is to keep advisors in front of learners instead of in front of paperwork. Every workflow above is an advisor doing that translation themselves, one tab at a time. They shouldn't have to.
What's the gap actually costing your program?
If your advisors are stitching together their own AI workflows to do the job, the gap between what your system gives them and what they actually need has a number on it. Most programs have never tried to calculate it.
Takes about 2 minutes. No demo, no sales call, no email required.
Tolu founded Clarivue after years as an academic advisor, watching qualified people lose jobs they were ready for - not because of skill, but because of how they performed in the interview room. He works with workforce development organizations and training institutes across Canada to help them scale interview preparation without scaling headcount.
