Case studyRecruitment ConsultingLondon / Remote
Computer Vision / Deep Learning Engineer
A role the client didn't believe could be filled — closed in 60 days.
A London-based deep tech startup building real-time AI anonymization technology
came to us with a position they considered nearly impossible to fill. We closed it in
60 days.
Candidates interviewed
12
by meet minds
Shortlisted
6
presented to client
Industry
Deep Tech
AI Privacy
the challenge
Computer Vision / Deep Learning Engineer
Why this hire was considered impossible.
The client arrived skeptical. They were building real-time, lossless anonymization technology for
machine vision a domain that sits at the intersection of generative AI, computer vision, and
privacy engineering. The role required hands-on experience with diffusion models, GANs, and
autoencoders, combined with strong software engineering in Python and C/C++. A PhD or MSc
was strongly preferred.
This isn't a standard AI hire. The pool of candidates who can both research state-of-the-art CV
models and ship production-grade code in this specific domain is extremely small globally and
the client knew it.
Ultra-niche technical profile
Generative models expertise (Diffusion, GANs, Autoencoders) combined with applied computer vision a rare combination even in top-tier AI markets.
Research and engineering in one profile
The role required someone capable of engaging with cutting-edge academic research and translating it into working systems. Most candidates excel at one or the other.
Academic credentials required
PhD or MSc preference meant the talent pool was anchored in academia a market most recruiters don't reach through conventional channels.
how we ran it
Our approach.
We staffed the search with consultants who understand both the academic AI landscape and the
deep tech hiring market. Sourcing was entirely outbound no job boards. We mapped the qualified
candidate pool directly, prioritizing researchers and engineers with proven experience in
generative models applied to visual data.
Every candidate was rigorously qualified by meet minds before reaching the client. Of the 12
professionals we interviewed internally, 6 met the bar and were presented. The client's time was
protected throughout.
process
How the search unfolded.
1
WEEK 1–2
Brief & calibration
Deep technical brief with the hiring team. We built a precise scorecard covering generative modelling depth, CV application experience, and engineering proficiency. Target profile agreed before any outreach began.
2
WEEKS 3–6
Active outbound & internal screening
Outbound-only sourcing across academic networks, AI research communities, and deep tech talent pools. 12 candidates interviewed by meet minds. 6 cleared the bar and were presented to the client.
3
FINAL PHASE
Client interviews & close
Structured client-side interview process, offer support, and signed contract. Role closed at day 60.
outcome
Result
Expert CV / DL Engineer — signed in 60 days.
A role the client had low confidence in filling was closed with a qualified, PhD-level candidate
sourced through outbound research. Six shortlisted profiles. Zero job board posts.
faq
common questions
Why 60 days for a role this specialized?
Because the qualified pool is genuinely small. We ran an outbound-led process with rigorous internal screening not volume. 60 days for an ultra-niche deep learning research role is a strong outcome. Rushing with unqualified candidates would have cost more time.
Do you handle similar deep tech mandates?
Yes. We work regularly with AI, deep tech, and research-heavy companies where conventional recruitment methods fall short. Brief us and we'll come back within 48 hours with a feasibility view.
Can you guarantee a similar timeline?
Every search has its own dynamics. We commit to a realistic timeline after the calibration phase and report against it throughout the process.
You presented 6 candidates did the client interview all of them?
Yes, the client interviewed all 6. That's by design. Every candidate we present has already passed a rigorous internal screening by meet minds we don't forward profiles that haven't cleared our own bar first. When a candidate reaches the client, they're genuinely worth the time. The client doesn't sift through volume; they meet qualified people.
more stories