Skills Assessment AI Draws $5M as Hiring Proof Shifts

Hiring on a small team means guessing whether someone can actually do the work, and a startup just raised money to remove that guess. On July 16, DeweyLearn announced in a PR Newswire press release that they raised a Series A funding round of $5 million. The oversubscribed round, as noted in the press release, was “led by SJF Ventures with participation from Catalysis Capital, Morningside, and Owl Ventures, among others.”

If you have ever hired someone with a flawless resume who could not do the job, you already understand the problem this addresses. Evaluating skill properly takes hours you do not have, so most small teams shortcut it and absorb the cost later. That shortcut is getting harder to justify.

Inside the $5 Million Round and Who Backed It

DeweyLearn builds multimodal AI that reads the visual, audio, and contextual information generated during a learning activity. Rather than logging what someone clicked or selected, it evaluates how the work was performed.

As noted by the above press release by PR Newswire, the company works “across clinical and healthcare education, higher education, workforce learning, and K-12.” That range matters, because assessment methods usually break when they move between hands-on and knowledge-based fields.

One deployment shows the scale. At the Auguste Escoffier School of Culinary Arts, the system was involved in grading over 20,000 submissions made by students, giving instructors an initial review, and returning feedback to students faster than before. Arrun Kapoor of SJF Ventures framed the appeal around better learning outcomes and career access, pointing to the founders’ combined industry and AI background.

Why Credentials Stopped Telling You Much

Degrees and certificates were always proxies, and the proxy is weakening. Skills shift faster than curricula update, so a credential earned three years ago describes a snapshot rather than a current capability.

Employers feel this directly. Research from Robert Half found that 62% of hiring managers said skills gaps had worsened over the previous twelve months, while only 6% said they had improved.

The frustrating part is that spending has not fixed it. Companies keep funding programs while employees still feel stuck, which is the tension behind so much workplace training that produces certificates instead of capability.

What hiring managers report about skills gaps

Response	Share of hiring managers
Skills gaps have worsened in the past year	62%
Skills gaps have improved	6%

Testing for Skill Without Building a Whole Process

You do not need enterprise software to do this better. You need one realistic task that resembles the actual job, scored against criteria you wrote down before anyone applied.

Keep it short and pay for it if it takes real time. A ninety minute exercise tells you more than three interviews, and compensating candidates keeps good people from dropping out of your process.

Write the scoring criteria first, though. Deciding what good looks like after you have seen the work invites you to rationalize a candidate you already liked, which is how bias enters a process that felt objective.

This applies well beyond engineering roles. Founders trying to scale sales often discover that a short call simulation predicts performance far better than a polished interview ever did.

What Faster Feedback Changes on a Small Team

Speed is the underrated benefit. The Escoffier example is about assessment volume, but the real gain is that feedback arrives while the work is still fresh in someone’s mind.

Small teams can copy that without any technology. Review work weekly instead of quarterly, and keep the feedback specific to a task rather than a personality. People improve against clear examples and stall against vague impressions.

It also changes who you can hire. When you can verify skill directly, you stop filtering for pedigree and start finding capable people others overlooked, which is how many Gen Z entrepreneurs ended up building outside traditional paths in the first place.

Hiring and Skills Questions Small Teams Face

Are skills tests fair to candidates? They can be more fair than interviews, provided the task mirrors real work and everyone is scored the same way. Problems arise when the exercise tests something the job never requires.

How long should an assessment take? Under two hours for most roles. Longer exercises filter for free time rather than ability, which quietly excludes people with caregiving responsibilities or a current job.

Should AI grade the results? As a first pass, potentially. A person should still make the decision, since the tool narrows the field while judgment about fit stays yours.

Hiring will not stop being hard. But replacing one interview with one realistic task is a small change that spares you the far more painful work of correcting a bad hire six months in.

The post Skills Assessment AI Draws $5M as Hiring Proof Shifts appeared first on Under30CEO.