How to spot fraudulent candidates before they waste your team's time
1 in 4 job candidates may be fraudulent by 2028.
Gartner
Fraudulent candidates are becoming a real problem in hiring, especially for remote technical roles.
If you've worked on engineering, AI, DevOps, or other hard-to-fill roles lately, you've probably felt it already. The resume looks great. The profile checks enough boxes to make you curious. Then the interview starts and something feels off. Maybe they're reading answers. Maybe they won't turn their camera on. Maybe the LinkedIn profile looks brand new for someone who supposedly has ten years of experience. Or maybe the person who shows up later doesn't match the person you screened.
At HighlightTA, we've been talking about this more and more with our team because it's no longer a one-off issue. It's a pattern. And while there's no perfect way to eliminate every fraudulent candidate, there are absolutely ways to spot the signs earlier and tighten up your process.
The good news is that most fake candidates do leave clues. You just need a consistent way to catch them.
This is especially true for growing tech companies that don't have a dedicated talent acquisition team. When hiring managers or small HR teams are running screens on their own, there's often no structured process to catch these issues early. That's a big part of what we help with at HighlightTA.
Why this is getting harder
Hiring teams are dealing with a few challenges all at once right now.
AI makes it easier to create polished resumes, over-rehearsed answers, and profiles that look convincing at first glance. According to a 2025 report from Career Group Companies, 65% of job candidates are now using AI at some point in the application process. That means resumes are more polished, answers are more rehearsed, and it's harder to tell who actually did the work. One-click applying has made inbound volume harder to manage. And for remote roles, especially technical ones, it can be difficult to confirm that the person interviewing is actually the person who would be doing the job.
Remote hiring makes this even harder. Without in-person touchpoints like ID checks or office based onboarding, remote roles remove layers of verification that used to happen naturally. For companies hiring distributed teams across Canada or the US without a dedicated TA function, this is a real blind spot.
That doesn't mean inbound hiring is broken. It just means recruiters need better filters.
We've found that the answer is not paranoia for the sake of paranoia. It's process. A little more structure upfront can save a huge amount of wasted time later.
“65% of candidates are now using AI at some point in the application process”
Where fraudulent candidates usually reveal themselves
We've seen the biggest tells show up in four places:
1. Resume and application review
2. LinkedIn and online presence
3. Screening interviews
4. Offer and onboarding steps
Here's what to watch for at each stage.
1. Resume review: the first filter
Sometimes the first signs are right there in the application.
Strange email patterns
One of the quickest early tells can be the email address. Things like random strings of numbers, unusual add-ons, or formats that feel mass-created can be a signal that something is off. On its own, this does not prove fraud. But paired with other concerns, it should put you on alert.
Over-polished resumes
Another common pattern is a resume that looks almost too perfect. Everything is bolded. The wording is polished to death. The skills are broad and impressive, but somehow still vague. It reads more like it was engineered to match search terms than written by a real person.
This is where recruiters need to trust their instincts. A great resume should still sound human.
Profiles that do not align
If the application says one thing and the public profile says another, pause. We've seen cases where the resume is tailored for a technical role, but the LinkedIn profile tells a completely different story. Sometimes it's sloppy. Sometimes it's a sign the candidate is using generated materials and forgot to update everything else.
Either way, mismatch matters.
2. LinkedIn should be part of your process
If we had to pick one easy upgrade to most hiring processes, this would be it: make LinkedIn review part of the shortlist process.
We've seen what happens when it isn't. On a Financial Analyst search, back-to-back applicants came in with the exact same resume format, only the name changed. Everything cross-checked on paper. But when we went to click the LinkedIn URLs, they were broken. Or the profiles weren't there at all. That's when it stopped being optional for us.
Not necessarily for every applicant in the first sweep, especially when volume is high. But before scheduling, it's worth doing a quick verification pass.
Here are some common LinkedIn red flags:
Very few connections
If someone claims years of experience in a technical field but has almost no network, it is worth a closer look.
Little to no activity history
A profile that only seems active in the last few months can be a warning sign. We've seen examples where profiles were built up quickly to look credible, but there's no real history behind them.
Missing photo or broken link
Not everyone is active on LinkedIn, but in many industries, especially for experienced talent, a completely empty or broken profile creates friction. If the link doesn't work, ask for the correct one. If they can't provide it, that tells you something too.
No verification badge
LinkedIn now lets users verify things like their workplace and identity. If a candidate claims years of experience at well known companies but hasn't verified any of them, that's worth noting. It's not proof of fraud on its own, but verified profiles do add a layer of trust. And when everything else is borderline, it can be the detail that tips the scale.
Look beyond the candidate's profile
Sometimes the issue is not the candidate. It's the companies listed on their profile. It's worth clicking through to those pages, not just reading the names.
We've caught this on a DevOps search. A candidate's profile looked solid on the surface, but the company page told a different story: wall-to-wall DevOps engineers, no finance, no HR, no leadership. That detail saved us from a bad hire, and checking company pages has been part of our standard LinkedIn review ever since.
A fast LinkedIn review can save you from a lot of wasted interviews.
3. Screening interviews: trust the signals
The recruiter screen is still one of the best fraud filters in hiring.
A good recruiter can spot things that never show up on paper. Tone, confidence, consistency, curiosity, genuine motivation. All the human stuff.
And if something feels off during the screen, it usually is.
Voice-only interviews can create risk
One of the simplest takeaways from our team discussion was this: when possible, use video.
If you never see the person during the process, it becomes much harder to know who is actually showing up later. This is especially important for remote technical hiring, where proxy interviewing has become more common.
That does not mean every camera issue is a red flag. But if a candidate avoids video throughout the process, you lose a valuable layer of verification.
Watch for muting and background coaching
A tell we've seen more than once is background noise followed by muting every time the candidate stops speaking. Sometimes it can sound like they are in a busy room or being fed answers. Again, one signal alone is not enough. But patterns matter.
Watch their eyes and body language
Video interviews give you more than just a face to match to a name. Pay attention to eye contact. Are they reading from another screen? When you finish asking a question, do their eyes dart to a different direction before they start answering? Do they consistently ask you to reword questions or slow things down? None of these are automatic disqualifiers, but when you see a few of them stacking up in the same conversation, it's a signal worth paying attention to.
Listen for answers that sound good but say nothing
This is probably the biggest one.
Fraudulent candidates often have polished answers to generic questions. But the moment you go off script, the quality drops.
That's why better questions matter.
4. The questions that can expose a fake candidate
Generic questions get generic answers. If you want to know whether someone actually did the work they say they did, you need to ask things that require real ownership, real memory, and real emotion.
But asking the right question is only half of it. The follow up is where the real signal lives.
A few that stood out from the conversation:
"What was your biggest contribution to your team?"
This is a great question because it pushes past buzzwords. Strong candidates can usually talk about what they actually changed, improved, built, or influenced. They can explain impact. They can tell you what they owned. They often bring in details or numbers naturally.
If someone can only answer in vague generalities, that is useful information.
Then follow up. If they mention a specific tool or approach, ask why they chose it. If they talk about a result, ask what didn't work along the way. Real candidates can go deeper because they lived it. Fraudulent candidates tend to stall or circle back to the same surface level answer.
"Why are you looking for a new opportunity right now?"
Real candidates usually have context. Sometimes too much context. They'll talk about growth, leadership, culture, change, burnout, or a contract ending. They sound like people with a real work history.
Fraudulent candidates often default to safe, generic lines like wanting a new challenge without giving you much underneath it.
"What do you love about your work?"
This is such a simple question, but it's powerful. People who genuinely do the work usually light up a bit here. Developers talk about projects. Salespeople talk about the thrill of closing. Good candidates sound connected to their craft.
That emotional layer is hard to fake consistently.
This is also where follow ups matter most. If someone says they love solving complex problems, ask them to walk you through one. Try and tie it back to emotion rather than just straight answers. Does their energy match what they're saying? Does the follow up align as a logical continuation of the first answer? That's where you learn the most.
"Have you worked on any personal projects?"
For technical candidates, this one is especially useful. GitHub contributions, side projects, experiments, things they've built for fun or learning. Those answers are usually much easier for a real candidate than for someone relying on scripts.
Even if they don't have personal projects, how they answer can still tell you a lot.
“41% of IT, cybersecurity, risk, and fraud leaders say their company has hired and onboarded a fraudulent candidate”
Technical roles need extra protection
Suggested section image: an engineering interview with code on screen and interviewers watching collaboration in real time
This issue came up again and again in the discussion: technical roles are where the problem feels worst.
That makes sense. The market is crowded, the resumes can be keyword-heavy, and it's easier for someone to sound capable in an early conversation than to prove it in a real working session.
And it's getting worse. CodeSignal found in early 2026 that cheating on technical assessments had doubled in just one year, from 16% to 35%. That's a huge jump and it shows why live assessments matter more than ever.
A few ways to reduce risk here:
Use live technical assessment formats
Paired exercises, live problem solving, and collaborative technical interviews make it much harder for someone to hide behind prepared answers.
We've seen what avoidance looks like. A candidate kept finding reasons the live coding session couldn't happen. A reschedule, an excuse, a request to swap it for something else. Eventually they refused outright. We held the line and they withdrew. A candidate who has genuinely done the work doesn't need to avoid showing it.
Rethink how you assess AI use
Here's an important one. The way coding is going, most code won't be written by humans. AI is already a massive part of how developers work, and that's only going to increase. So the question shouldn't be "are they using AI?" It should be "do they know how to use it well?"
You don't want to eliminate candidates for using AI in a technical assessment. That would be like penalising someone for using Google ten years ago. What you want to understand is whether they have the knowledge to guide the AI, not just get guided by it. Can they explain the logic behind the output? Can they spot when it's wrong? Can they talk about the project, the trade-offs, and the decisions they made along the way?
That's the difference between someone who understands the work and someone who's just copying and pasting outputs.
Prep interviewers on what to watch for
Fraud detection should not sit only with the recruiter. Hiring managers and technical interviewers should know the signs too, especially if they're the ones most likely to notice when a candidate is reading from AI or struggling to explain their own reasoning.
Compare communication across stages
Did the candidate sound one way in the first round and completely different in the next? Did their level of fluency or confidence swing wildly? Those inconsistencies are worth discussing internally before moving ahead.
“Cheating on technical assessments doubled in one year, from 16% to 35%”
Offers and onboarding: don't drop your guard too early
Suggested section image: a recruiter delivering a verbal offer on video to confirm candidate identity
One of the trickiest parts of this problem is that sometimes concerns only show up after offer stage.
For example, a candidate signs the offer, then suddenly starts asking unusual logistics questions. They want equipment shipped somewhere other than their home address. They introduce a last-minute travel situation that changes basic onboarding steps. On their own, these things may be explainable. But sometimes they're the first visible crack.
This is where policy helps.
Use policy instead of personal judgment
If equipment only ships to a home address, say that clearly. If the candidate pushes back hard or can no longer proceed, that may tell you what you need to know.
Policies take emotion out of the conversation and make it easier for recruiters to hold the line.
Consider verbal offers on video
A short video call at offer stage can be a helpful check, especially if it refreshes your memory of who the candidate is before onboarding begins.
Keep HR or another stakeholder involved when needed
Sometimes a second set of eyes helps. If there is any uncertainty, involving another internal stakeholder can give you an additional layer of confidence.
Build a checklist, not a guessing game
The biggest takeaway from all of this is that we probably do not need one magic solution. We need a repeatable process.
A good anti-fraud checklist might include:
Resume review checklist: Look for formatting oddities, inconsistent experience, suspicious email patterns, and overly generic language.
LinkedIn review checklist: Check connections, activity history, profile quality, verification badges, company legitimacy, and whether the profile aligns with the resume.
Interview checklist: Use video where possible. Ask deeper questions. Use follow ups to test depth. Watch for scripted answers, eye movement, inconsistencies, and unusual interview behavior.
Offer and onboarding checklist: Lean on policy. Confirm logistics. Be cautious when new issues suddenly appear after the offer is signed.
When these checks are built into the workflow, recruiters do not have to rely on memory or instinct alone.
If you're using Ashby as your ATS, it's worth knowing they launched a Fraudulent Candidate Detection feature that runs checks automatically when candidates apply. It flags suspicious applications in the background without adding friction for real candidates. We use Ashby across our client work and it's one more layer built into the process.
Final thoughts
Fraudulent candidates are a real challenge, and they're not going away anytime soon.
But this does not mean your team has to accept wasted time as part of the job. Most fake candidates are not perfect. They leave clues in their resumes, their LinkedIn profiles, their interview behavior, and the questions they struggle to answer.
The key is consistency.
A solid process helps recruiters move faster, protect hiring managers' time, and create a better experience for the real candidates you actually want to hire. Especially in technical hiring, where the volume of questionable profiles can feel exhausting, a few extra checkpoints can make a huge difference.
At HighlightTA, we're big believers that the best hiring processes balance speed with good judgment. This is one of those areas where a little extra structure upfront can save a lot of pain later.
And honestly, that's the goal. Not to become paranoid. Just to get sharper.
If you don't have a TA team catching this stuff for you, that's exactly what we do. HighlightTA plugs into your team with the tools, process, and experience to filter out the noise and protect your hiring quality. Whether you need embedded talent partners or a full recruitment function built from scratch, we can help. Get in touch to learn more.