👋 Hi, I’m Andre and welcome to my newsletter Data Driven VC which is all about becoming a better investor with Data & AI. ICYMI, check out some of our most read episodes:
Branding, thought-leadership, lead-gen, activation - we got you covered. Existing partners include Affinity, AlphaSense, Carta, Crunchbase, Harmonic, Hubspot, HSBC Innovation Banking, Specter, Vestberry, and many more.
We’ve just opened 2026 slots and are happy to explore new partnerships.
Welcome back to another Data Driven VC “Insights” episode where we cut through noise, scrutinize the latest research, and translate complex findings into practical, evidence-based takeaways for investors, founders, and operators.
This newsletter is a lot about leveraging AI for workflow automation and enhancing our decision-making with data-driven processes. While we focus primarily on the first-order implications, it’s equally important to look at second-order effects in our portfolio.
AI is rapidly reshaping how startups hire, but without redesigned processes, better measurement, and trust-building, most teams are still losing top talent before the first interview even happens.
Today, we uncover how the best teams hire in the age of AI, what talent they’re looking for and how best-in-class processes convert the best candidates.
✅ TL;DR (5 Key Takeaways)
Generative AI has shifted hiring demand toward AI, ML, and data roles, while entry-level hiring has collapsed, intensifying competition for senior talent and weakening long-term pipelines.
Startups lose most candidates early in the funnel, often due to slow responses, clunky applications, scheduling friction, and premature automated screens.
Candidate trust is becoming a differentiator, as opaque AI usage and lack of human interaction increasingly turn high-quality candidates away.
AI improves hiring only when it reduces friction and increases speed and clarity for candidates, not when it automates broken processes.
The highest-performing teams treat hiring as a measurable system, tracking funnel metrics weekly and optimizing for long-term quality, retention, and experience rather than just speed.
Why This Matters
More AI recruiting tools than ever exist, yet candidate drop-off before the first interview remains stubbornly high, with some studies estimating drop-off rates as high as 80% (TalentAlly, 2025).
Hiring experience matters enormously: 78% of candidates in LinkedIn industry research said the hiring experience is a relevant indicator of how a company treats employees, and candidates with negative experiences are likely to share this with their professional network, damaging your reputation (Tellent Recruitee, 2025). The goal, then, is not simply to adopt tools, it’s to fundamentally improve hiring processes so that top talent feels valued from the moment they apply.
The irony is stark: AI was supposed to speed hiring and reduce friction. Instead, many startups have created friction machines that lose candidates at precisely the moment they should be converting them (Talentpool, 2023). The real problem isn’t the tools themselves, it’s that most teams bolt AI onto broken processes without redesigning hiring around how candidates actually move through funnels.
Before optimizing with technology, you need to first understand and fix the underlying process (A Job Thing, 2024 a).
The New Talent Market: AI Is Reshaping Which Roles Get Hired
The composition of hiring has shifted sharply in the past 18 months, a change driven almost entirely by generative AI adoption. AI and ML, data, and platform roles are exploding as companies rush to hire professionals with AI-specific skills (Lightcast, 2025). Lightcast's 2025 analysis of the generative AI job market shows significant growth in AI-specialized roles in 2024-2025, representing one of the fastest-growing job categories in tech (Lightcast, 2025).
Simultaneously, traditional support, operations, and admin roles are shrinking (Ravio, 2025), not necessarily because AI is perfect at these tasks, but because companies believe AI can handle them, leading to underinvestment in human headcount.
The most striking trend is the collapse in entry-level hiring. SignalFire data shows entry-level hires (≤1 year experience) fell from approximately 15% to 7% of total hires at VC-backed startups since 2019, a 50% decline (SignalFire, 2025).
This has profound consequences: fierce competition for mid- and senior-level AI talent intensifies, while the traditional on-ramp for new entrants, career switchers, and non-traditional backgrounds is systematically closing (SignalFire, 2025). The market is bifurcating into two tiers: scarce, expensive senior AI talent versus struggling junior pipelines (BCG, 2025).
This trajectory will create serious ecosystem challenges. A weaker leadership pipeline means fewer future CTOs, staff engineers, and founding team members in 5 to 10 years (Ravio, 2025). Moreover, juniors are often the first to experiment with new AI tools and workflows; seniors typically stick with they already know. Without junior employees, organizations under-adopt AI or centralize it in a single “AI expert”, which is fragile and slow. The assumption that junior work can be automated entirely is dangerously short-sighted (Mohamed Yasser, 2025).
Instead of eliminating junior roles, smart companies should redesign them: create AI-native junior roles where a junior plus an AI stack can handle work that previously required a mid-level engineer or coordinator.
✈️ KEY INSIGHTS
GenAI has rapidly reshaped hiring, driving strong demand for AI, ML, and data roles while reducing investment in traditional support and operations positions. At the same time, entry-level hiring at VC-backed startups has collapsed, intensifying competition for senior AI talent and weakening long-term talent pipelines.
Rather than removing junior roles altogether, forward-looking companies will need to rethink them by pairing early-career hires with AI tools so they can deliver impact previously expected from more experienced staff.
Where Startups Actually Lose Candidates
Most hiring advice focuses on interview technique, compensation negotiation, or offer structuring. But the real leak happens much earlier: candidates ghost or withdraw before you ever speak with them (The Interview Guys, 2025; TalentAlly, 2025). Research into candidate drop-off patterns identifies discrete funnel stages where talent disappears: job view to application start, application start to completion, completed application to first human response, and from first contact to scheduled interview (A Job Thing, 2024 a).
The bottleneck varies by company, but common failure modes are consistent:
Slow or absent first response signals disorganization to top candidates who have options elsewhere (TalentAlly, 2025);
Clunky application forms or ATS processes cause candidates to abandon mid-form (Tellent Recruitee, 2025);
Scheduling friction (endless calendar coordination, no timezone awareness, vague agendas) creates friction that high-signal candidates simply won't tolerate (Tellent Recruitee, 2025);
Pre-screen assessments or "quick tests" before a conversation signal disrespect for their time. Each of these failure modes is fixable, but only if you first measure where the leak is largest (TalentAlly, 2025).
Companies that respond quickly and consistently to promising candidates tend to see higher conversion to first interview and better offer‑acceptance rates, although the exact impact varies by role and market (Dover, 2025). Candidates with multiple offers don't wait; they move forward with whichever company feels most organized and respectful of their time. Your AI tools are irrelevant if they enable slow, impersonal, template-driven outreach that high-quality candidates recognize and ignore.
Candidate Trust in an AI-Heavy Hiring Process
Top candidates are increasingly skeptical of AI in hiring. They worry about black-box screening that rejects them without explanation, biased training data in AI models that may discriminate based on background or demographics, and a loss of human connection that signals the company doesn't care about them as individuals (Newcastle University, 2025).
Yet most startups don't explicitly address AI in their hiring messaging or process. This creates a trust gap, especially with the most discerning candidates, exactly the ones you most want to hire (The Talent Bulletin, 2025).
How do you rebuild trust?
Be transparent about where and how AI is used in your process. Candidates who understand that AI is supporting workflow while humans make final decisions report higher levels of perceived fairness and trust (Hirebee, 2025).
Offer feedback loops and a path to human review when automated systems screen people out, since lack of explanation is a major driver of negative candidate experience and reputational damage.
Keep humans visibly involved in communication: personalized outreach and clear expectations from a named recruiter or hiring manager mitigate the “black box” feeling that turns senior candidates off (HBR, 2024; Newcastle University, 2025).
Use AI only where the benefit to candidates is obvious. Faster scheduling, fewer emails, clear logistics, and timely updates – all core elements of a strong candidate experience (Personio, 2025).
Developing an Inclusive and Sustainable AI in the Human Resources Management Ecosystem (Newcastle University, 2025)
✈️ KEY INSIGHTS
Startups tend to lose candidates long before interviews begin, mainly due to slow responses, clunky application processes, scheduling friction, and early automated screens that signal low respect for candidates’ time. At the same time, growing skepticism toward AI-driven hiring creates a trust gap when companies fail to explain how automation is used or to keep humans visibly involved.
The core takeaway is that speed, clarity, and transparency matter more than tooling, and AI only improves hiring outcomes when it demonstrably reduces friction and reinforces trust rather than replacing human judgment.
How to Act On This
Most hiring problems are not caused by a lack of tools but by a lack of measurement, review evidence, and accountability across the talent funnel.
Build a talent funnel dashboard, not just a recruiting stack:
Installing ATS or AI tools does not improve hiring by default. Founders should define and track a small set of funnel KPIs that expose where candidates drop off, including time to first response, application completion, conversion rates between stages, offer acceptance, and total time to close (A Job Thing, 2024 b)
Set clear benchmark targets for every funnel stage:
Establish explicit targets such as sub-24-hour response times for priority candidates, application completion rates above 75%, and offer acceptance above 70%. When metrics fall below benchmarks, treat it as a system failure, not a sourcing problem (Dover, 2025).
Break funnel metrics down by role, seniority, and source:
Review hiring funnel data weekly and fix leaks before adding volume:
A weekly funnel review allows teams to spot ghosting, scheduling friction, or assessment drop-off early. When a stage underperforms, diagnose and test improvements rather than compensating by sourcing more candidates (Dover, 2025)
Optimize for long-term quality, not just hiring speed:
Track outcomes beyond the hire, including one-year retention, early promotions, candidate experience feedback, and cohort diversity. This reframes hiring as an evolving system that compounds performance over time rather than a transactional process (A Job Thing, 2024 b).
✈️ KEY INSIGHTS
Most hiring failures stem from missing measurement and accountability, not from a lack of ATS or AI tools. Founders who define clear funnel benchmarks, segment metrics by role and source, and review them weekly can identify and fix candidate drop-off before adding more volume. The strongest teams optimize hiring as a system, tracking long-term quality and retention rather than just speed to hire.
Thanks to Lea Winkler for her help with this post.
Stay driven,
Andre
