Vena Medical Case Study

The Client

Company	Vena Medical Holdings Corp.
Industry	Medical Devices - Intravascular Imaging & Catheter Technology
Company Size	~17 employees (Y Combinator W18)
Location	Kitchener, Ontario (Medical Innovation Xchange)
Role Hired	R&D and Manufacturing Catheter Assembler
Virvell Modules	AI Pre-Screen Interviews + Background Verification
Recent Milestones	FDA clearance, $4.5M Manufacturing Centre of Excellence, 100+ patients treated

The Challenge

500 Applications, Zero Bandwidth

Vena Medical was at an inflection point. They had just received FDA clearance for their MicroAngioscope System - the world's smallest intravascular camera - opened a $4.5M Manufacturing Centre of Excellence, and were preparing to enter the U.S. market. Ontario's Minister of Economic Development cut the ribbon at their new facility.

To scale production, they needed to hire a specialized R&D and Manufacturing Catheter Assembler - someone with cleanroom experience, GMP documentation skills, and ideally hands-on catheter assembly background. It's a niche role in a competitive corridor (Kitchener-Waterloo) alongside employers like Baylis Medical, StarFish Medical, and Medtronic.

They'd tried this before - and it didn't work. CEO Michael Phillips described what happened the last time they posted on Indeed: within three days, 500 applications flooded in, and there was no way to filter them. With no HR department, no recruiter, and two engineer-founders juggling FDA submissions and manufacturing setup, the applications sat unreviewed.

This time, when the posting went live across Dover, Indeed, and LinkedIn, 123 applications arrived within days. But this time, they had Virvell.

The core problem:

500 applications sat unreviewed last time. 123 new applications arrived this time. A 17-person medtech startup with no HR function needed a way to screen every applicant consistently for a highly specialized manufacturing role - without pulling engineer-founders away from FDA clearance and U.S. market entry.

The Solution

What Virvell Did

Through a consulting engagement via the Medical Innovation Xchange (MIX), Virvell deployed its AI-powered screening platform to handle the entire pre-screening process for Vena Medical's catheter assembler hire.

1

Built a Bill 149-Compliant Job Posting

Ontario's Working for Workers Act requires AI disclosure, pay transparency, and vacancy status on all job postings. Virvell ensured the Dover careers page and Indeed posting included all required disclosures from day one - including the explicit statement that AI assists in screening while a human reviewer makes all hiring decisions.

2

Configured Role-Specific AI Pre-Screening

The platform was configured with 12 custom pre-screening questions tailored to the catheter assembler role: cleanroom experience (ISO class), GMP documentation familiarity, catheter/guide wire assembly background, microscope and precision work, and availability/compensation alignment. Every candidate received the same structured evaluation - no variance between calls, no interviewer bias.

3

Screened 123 Candidates via Voice AI

Rather than cherry-picking resumes, Virvell ran AI-powered voice pre-screens on all 123 applicants. The platform called candidates directly and left voicemails for those who didn't answer, with a callback code so they could complete their pre-screen on their own time. Calls were available 24/7 — and candidates took advantage of it, with 53 calling back inbound on their own schedule, including 8 candidates who completed after 6 PM Toronto, one of them just past midnight.

4

Delivered Structured Reports in 26 Seconds

Each completed pre-screen automatically generated a structured report with the candidate's responses, skills verification, red flag detection, and an AI recommendation (proceed, review, or hold). Reports were generated in an average of 25.7 seconds after each call (median 21 seconds, 90th percentile 49 seconds). Every completed pre-screen produced a structured report — no manual follow-up required.

The Results

78 Screens. 44 in the First 3 Hours. Zero Client Hours.

Metric	Result
Total Applications	123
Pre-Screens Completed	78 (63%)
Completed in First 3 Hours	44 candidates
Peak Hour	1 PM EDT, March 11
Peak Concurrent Calls	6 simultaneous (instantaneous peak at 1:47 PM)
Sustained ≥4 Concurrent	34 distinct minutes
AI Recommendation: Proceed	5 (6%)
AI Recommendation: Review	42 (54%)
AI Recommendation: Hold	31 (40%)
Inbound Calls	53
Inbound Completion Rate	98.1% (52 of 53)
Outbound Calls	70
Outbound Completion Rate	37.1% (26 of 70)
Lift Multiple	2.6×
After-Hours Completions (after 6 PM)	8 (including one post-midnight)
Reports Generated	78 of 78
Avg Report Generation Time	25.7 seconds
Shortlisted for Interview	3
Hire Made	Yes
Client Hours Screening	0

How the AI Differentiated Candidates

The AI wasn't sorting randomly. Candidates who lacked the required experience couldn't answer the technical questions - the screening was self-selecting.

Proceed (5 · 6%)

Strong fit on verified skills, minimal red flags, full answers across the structured script. Direct experience in catheter assembly, regulated manufacturing environments, and the dexterity-related work the role required.

Review (42 · 54%)

Mixed signals — some skills verified, some gaps worth a closer look from the founders. Adjacent manufacturing experience but not catheter-specific, or the right technical background with shift-availability questions, or strong soft answers with one missing requirement.

Hold (31 · 40%)

Significant gaps in role-specific technical answers. Often unable to describe the manufacturing environments, equipment, or processes the role required.

The screening was self-selecting: the AI didn't reject Hold-tier candidates — they couldn't answer the questions because the experience wasn't there. The AI detected the difference.

The Timeline

Day One Throughput, Multi-Day Completion

On March 11, 2026, Virvell began screening candidates at 12:30 PM EDT. The peak hour ran from 1:00 to 2:00 PM EDT, with up to 6 candidates on parallel calls at 1:47 PM and the platform sustaining 4 or more concurrent calls for 34 distinct minutes. By 3:30 PM, 44 pre-screens were completed in the first three hours — what would take a human recruiter conducting 20-minute screens nearly three full workdays. Inbound callbacks continued through the evening and into the night, including one call that wrapped just past midnight. Callbacks continued for five more days, including weekends, ultimately reaching 78 completed screens.

The screening timeline:

12:30 PM EDT — Outbound calls begin reaching all 123 candidates

1:00–2:00 PM — Peak hour. Up to 6 simultaneous calls running, instantaneous peak at 1:47 PM

3:30 PM — 44 candidates completed in the first 3 hours

6:00 PM onward — Inbound callbacks accelerate as candidates wrap up their workday

Evening and overnight — 8 completions land after 6 PM across the pilot window, one wrapping just past midnight

Days 2–5: Callbacks continued through the week. Final tally: 78 completed.

Inclusion Arbitrage: When the Candidate Chooses the Moment

When candidates called back inbound on their own time, 98.1% completed the full screen (52 of 53), compared to 37.1% on outbound calls (26 of 70). A 2.6× lift. Of the 123 applicants, 53 reached the platform via an inbound call rather than answering an outbound — and nearly every single one of them completed the screen.

Eight of the 78 completed pre-screens landed after 6 PM Toronto. One wrapped just past midnight. These were candidates who couldn't take a recruiter's 2 PM call because they were on shift, in standups, picking up kids, or otherwise unavailable during traditional screening hours. No recruiter is available to take those calls. The AI was.

There's a name for what that does: Inclusion Arbitrage — capturing the talent pool the 9-to-5 model never reaches. The number is modest at single-pilot scale. The pattern is what matters: when you remove the scheduling tax with a 24/7 AI line, the candidates who self-select in are the ones the traditional model systematically excludes.

	Traditional	With Virvell
Time to screen 123	3–4 weeks	One day + 5-day callback tail
Client hours	30–40+	0
Consistency	Varies by call	Same structured questions for every candidate
After-hours availability	None (9–5)	24/7
Simultaneous calls	1 at a time	Up to 6 concurrent
Bill 149 compliance	Manual effort	Built-in
Comparison across candidates	Notes / memory	Side-by-side reports
Report speed	Hours / never	25.7 seconds avg

The Outcome

From 123 Applicants to 1 Hire

From 123 applicants, Virvell's AI pre-screening identified 5 "proceed" candidates and 42 for "review." Vena Medical's founders selected three finalists for on-site interviews and dexterity testing, drawing from across confidence tiers — validating the human-in-the-loop model. The AI didn't make the decision. It organized the information so the founders could.

The founders looked at the full AI reports, weighed the verified skills against the flagged gaps, and decided which candidates were worth meeting in person. That's the system working as designed: AI collects data, humans decide.

The hire:

Vena Medical made the hire. The successful candidate had received a "proceed" recommendation from the AI's underlying evaluation, with strong scores on verified skills and minimal flagged items.

Cross-Module Intelligence: 4 Facts Captured, 1 Flagged for Follow-Up

After the hiring decision was made, Virvell ran a background check on the successful candidate through its integrated Certn partnership. The background check returned clear with no criminal records.

Cross-Module Intelligence then extracted 4 verified facts from the candidate's pre-screen, identified 0 internal contradictions, and flagged 1 item (medium severity) for the hiring team's consideration during onboarding. The system surfaced what aligned, what didn't, and what needed a closer look — organizing the evidence the recruiter needs to make the call.

This is the platform working as designed. Pre-screen data feeding into a structured intelligence layer alongside background verification, with module-level source comparison. As more candidates flow through both modules in future pilots, the cross-source comparison surface area grows — the moat is the architecture, and the architecture compounds with use.

Cross-Module Intelligence output for the hired candidate:

4 verified facts. 0 contradictions detected. 1 item flagged for human review. Background check: all clear.

123 applied → 78 screened by AI → 5 proceed + 42 review → 3 interviewed → 1 hired

Total time from job posting to background-check clearance: 21 days. Client screening time: zero.

"Working with Julien and Virvell was great. Initially, we weren't sure what additional value we'd gain from working with an external HR consultant versus doing it internally, but the entire process was seamless and turnkey. We provided the role details and requirements, as well as some light input on the shortlist of candidates, and he managed the rest. By the time we interviewed, we had three high-quality candidates along with a custom interview guide that Virvell prepared for us. Julien guided us through the interview process all the way to the final offer. It was a tailored and efficient experience, and I'd gladly work with him again in the future."

Phillip Cooper, COO, Vena Medical

"We tried posting this job on Indeed, and we got, within three days, 500 applications. And I couldn't filter through them... it was just impossible."

Michael Phillips, CEO, Vena Medical

"Interviewing and hiring is a skill, one which I think we're still learning. It's a fair amount of mindshare managing interviews and the flow of everyone."

Phillip Cooper, COO, Vena Medical

Why This Matters

Vena Medical's story isn't unique. Across Canada, high-growth startups face the same dilemma: they post a role, get flooded with hundreds of applications, and have no way to process them. The default is the founder doing phone screens between investor calls and product launches - or, as Vena experienced firsthand, 500 applications sitting unreviewed because nobody had the bandwidth to filter them.

Virvell changes that equation. By combining AI pre-screen interviews, background verification, and voice AI reference checks into a single platform, companies like Vena Medical get enterprise-grade screening without needing an enterprise-sized team.

Key Takeaways

123 candidates screened, 78 completed, 1 hired — complete funnel from application to offer

44 pre-screens completed in the first 3 hours. A task that would take a recruiter most of a week.

Peak hour at 1 PM ran up to 6 simultaneous calls, with 4 or more concurrent for 34 distinct minutes

Candidates who called back inbound completed the screen at 98.1% versus 37.1% on outbound (2.6× lift). 53 candidates called back on their own time.

8 candidates completed pre-screens after 6 PM Toronto, including one just past midnight — the candidates the 9-to-5 phone screen would have missed

The candidate who was hired had received a "proceed" recommendation from the AI

Finalists were drawn from across confidence tiers — proving human-in-the-loop works

Cross-Module Intelligence on the hired candidate: 4 verified facts, 0 contradictions, 1 flag for human review

Bill 149 compliance built into the workflow, not bolted on after

AI collects data, humans decide — no candidate scoring, no auto-rejection

About Vena Medical

Vena Medical Holdings Corp. builds the world's smallest intravascular cameras, giving physicians a live view from inside blood vessels. Founded as a University of Waterloo capstone project in 2016 and backed by Y Combinator (W18), Vena Medical has treated 100+ patients across Canadian hospitals, holds 9 granted patents, and recently received their first FDA clearance for peripheral vascular use. They are headquartered in Kitchener, Ontario at the Medical Innovation Xchange.

From 500 Unfiltered Applications to a Structured Shortlist - and a Hire

Built a Bill 149-Compliant Job Posting

Configured Role-Specific AI Pre-Screening

Screened 123 Candidates via Voice AI

Delivered Structured Reports in 26 Seconds

About Vena Medical

Ready to screen smarter?