Everyone Wants AI Talent. Nobody Agrees What It Looks Like.
October 2026 · By Inbal Levy Horesh, Founder & CEO, Gifthead
Gifthead's Israeli Tech Hiring Report 2026
Israeli tech companies want AI people more than ever. They post dozens of AI jobs every week. And then they reject more than four in ten AI candidates before the first interview, twice the rate of other roles.
This is not only a talent shortage. It is a definition problem. Companies want AI talent, but they don't yet agree on what a good AI hire looks like.
We see it from the field, and now we can show it in numbers.
Key findings
R&D took over hiring. Nine in ten of the hires we closed in 2026 were R&D roles (90%), up from 54% in 2024.
Companies hire fewer people, and more senior ones. Six in ten 2026 hires (62%) were senior, staff, lead or management roles, up from about a third (35%) in 2024.
Israel hires a bit faster than the global market. About 70 days from opening a tech role to signing, vs. a global median of 75 days for tech roles. About three months until the new hire's first day.
The bar went up in AI roles. In 2026, 7 in 10 candidates we sent got an interview, down from 8 in 10 in 2025. Most of the drop comes from AI roles. In other roles, the rate barely changed.
AI roles are where hiring breaks. Clients rejected 44% of AI candidates on CV alone, twice the rate for other roles (22%).
All Gifthead numbers in this report come from our own hiring data (2022–2026). Market numbers come from public sources, listed at the end.
How we measured
This report draws on nearly 200 hires and more than 20,000 candidates we approached, screened and presented to clients between 2022 and 2026. Market data comes from public sources, listed at the end. All data is anonymous. No client or candidate is named.
Our clients are mostly startups and growth-stage tech companies, so our data leans toward R&D roles. The shifts are real in our data, but the exact percentages reflect our client mix, not the whole market. That's why we put our numbers next to the national and global data.
What Israeli tech companies hired in 2026
In 2026, hiring narrowed to one profile: the senior engineer who can work alone. Everything else shrank.
Share of Gifthead hires | 2024 | 2025 | 2026 (Jan–Sept) |
R&D (engineering, data, AI, tech leadership) | 54% | 78% | 90% |
Sales and marketing | 43% | 4% | 7% |
Other (product, design, operations, finance) | 3% | 18% | 3% |
Total | 100% | 100% | 100% |
Seniority moved the same way: senior, staff, lead and management roles were 35% of our hires in 2024, 46% in 2025 and 62% in 2026.
R&D became almost all of the hiring. In 2024, R&D was about half of our hires. In 2026, nine in ten.
We don't claim AI caused this alone. The market data says the same thing more carefully: AI's impact shows up mainly at the hiring stage, not in layoffs. When a company decides which roles it can delay, R&D is the last to wait. We see that decision being made in real time.
An old role got a new name. Early-stage companies always needed the one engineer who builds the first version of everything. In 2026 the role got its own title: "Founding Engineer" was about one in ten of our hires.
How long hiring takes
Most tech roles take 6 to 15 weeks from kickoff to signing, then 2 to 6 weeks until the first day at work. The time to signing barely moved since 2024. The wait for the first day grew, from about 3 weeks in 2024 to about 4.5 weeks in 2026.
That's a bit faster than the global pace. Ashby's 2026 data puts the median time from opening a technical job to the first hire at 75 days. Our median in 2026 was about 70.
This is the number founders forget when they plan. The hiring plan says "Q1". The new engineer writes code in Q2.
Where a role falls inside the range depends less on the market and more on the role: how senior it is, and how clear the brief is on day one.
The process got stricter, mostly for AI roles
In 2026, 7 in 10 candidates we sent got an interview with the client, down from 8 in 10 in 2025. Most of that drop comes from AI roles.
2025: wider door, later decision. Companies let more candidates into interviews, then said no later in the process.
2026: a tighter door for AI roles. Interviews stayed as strict as in 2025, and the door got tighter mainly for AI roles. In other roles, the share of candidates who got an interview barely changed. And this isn't just better screening. If it were, more of the interviewed candidates would move forward. The opposite happened: about 3 in 10 interviewed candidates moved forward in 2025, and about 1 in 4 in 2026.
From the field: same role, same pace, different decisions. One client hired for the same senior software engineering role in 2025 and 2026, with the same hiring need in both years. In both years, we sent them about two candidates a week. In 2025, about one in ten candidates we sent was hired: 11 hires. In the first half of 2026, one hire out of 38 candidates. The job didn't change, and the pace didn't change. What changed was the bar for a hire. It went up, the same way it went up across the market.
What didn't change: candidates say yes. Three in four offers were accepted in 2026. When a company finally decides, the right candidate usually takes the job. The bottleneck is not the candidate. It's the decision.
AI roles: where hiring breaks
For AI roles, our clients rejected more than four in ten candidates before talking to them. That's twice the rate of other roles.
2026, Gifthead data | AI roles | All other roles |
Candidates rejected on CV, before any interview | 44% | 22% |
Candidates who said "not interested" (of those who replied) | 73% | 65% |
AI roles = positions where building or applying AI and machine learning models is the core of the job, such as AI Engineer, ML Researcher, Agentic AI Team Lead and Data Scientist. Engineering roles that only use AI tools or APIs were counted as other roles.
The market says: there aren't enough AI people
That part is true. The Israel Innovation Authority and the Samuel Neaman Institute found an immediate shortage of about 2,400 "Talent" and "Top Talent" workers in core AI roles, mostly data scientists and data engineers. The government launched a program to bring AI experts to Israel because of it.
And demand is wide. The Neaman Institute tracks AI job ads on LinkedIn Israel every week: about 67 new AI developer ads a week, 3,601 ads from 914 different companies in one year.
Our data says: it's also a definition problem
Here is the part the job-ad data can't see. On paper, requirements are getting softer: in Neaman's data, the share of AI ads asking for a master's degree dropped from 13% to 7% in June 2026. In practice, it's the opposite. Our clients rejected 44% of AI candidates on CV alone.
Companies loosen the job description, then tighten when the CVs arrive. That's what it looks like when nobody agrees what the role is.
The titles tell the same story. Similar AI work goes by dozens of different names: AI Engineer, Applied AI Engineer, LLM Engineer, Agentic AI Engineer, Machine Learning Engineer, Algorithm Engineer, Data Scientist, Forward-Deployed Engineer, MLOps Engineer, AI Platform Engineer. Some have "AI" in the title and some don't. And two jobs with the same title can mean very different work.
And it's not only an Israeli problem. Ashby's 2026 data shows that when two interviewers score the same candidate, about 38% of pairs disagree, and nearly half of those disagreements cross the line between yes and no. Gem's 2026 benchmarks show Data Science now needs more interviews per hire than any other department (38.5).
A shortage you can fix by finding more people. A definition problem you fix before you start looking.
Israel vs. the world
On the two measures where we can compare like with like, Israeli tech hiring matches the global market in 2026.
2026 | Global (Ashby) | Gifthead, Israel |
Median time from opening a tech role to the first hire | 75 days | about 70 days |
Offers accepted by sourced tech candidates | 76% | 75% |
Ashby's data covers 54 million applications and 93,000 jobs worldwide, January 2021 to March 2026. For the wider picture, Gem's 2026 benchmarks find that 82% of all offers are accepted, the highest rate since 2021.
What this means if you're hiring now
1. Define the work and the decision before you open the role. Not the title. The work: what will this person actually do? Then decide how you'll test it, who interviews, who has the final say, and how the team reaches one decision together.
Here's why it's so hard right now: using AI at work is still new. Every CV says "works with AI", but there's no standard yet for what good AI work looks like, or how to test it. Each interviewer means something different by "AI experience", and so does every candidate, depending on where they come from. When the CTO, the team lead and the founder each picture a different hire, every candidate fails someone's test. That's how more than four in ten CVs get rejected before anyone talks to the candidate.
And remember: there may not be one title that fits. The right person may never have done this exact job before. Be creative in the interviews, and look for who can do the work well, not who has the matching line on their CV.
2. Plan for the first day, not the signing. About three months from kickoff to a person at the desk. If the roadmap needs them in Q2, the search starts in Q1.
3. Senior is a capability, not a title. Everyone is chasing the same "senior" title right now. What you actually need is the ability: owning a problem end to end, working without hand-holding, making good calls. Some engineers already do this without the title. Test for the ability, not the title or the years, and a much wider market opens up.
4. Don't read the layoff headlines as a hiring market. More job seekers doesn't mean easier hiring. In 2026, more people in Israel are looking for tech jobs, and companies still became more selective, especially in AI roles. The people you want are not the ones looking.
The companies that close AI roles fastest are not the ones with the biggest budget. They're the ones who knew what they were looking for before they started.
About Gifthead
Gifthead is a boutique tech recruitment firm in Israel. We bring strong, precise candidates, and build each client the hiring model and price that fit them. AI does the heavy lifting, so we can focus on what really matters: the people.
Founded in 2019 by Inbal Levy Horesh. Offices in Israel, the USA and Cyprus. gifthead.net
Sources
Ashby — Recruiting Operations Benchmarks, 2026 Talent Trends Report (time to first fill, offer acceptance by source, interviewer scorecard alignment)
Gem — 2026 Recruiting Benchmarks Report (offer acceptance, interviews per hire by department)
Israel Innovation Authority — Attracting Top AI Talent (AI talent shortage, with the Samuel Neaman Institute)
Israel Innovation Authority — Human Capital in Tech, H1 2026 (hiring, layoffs and AI's impact on hiring)
Samuel Neaman Institute — Trends in Demand for AI Developers in Israel, July 2026 (LinkedIn Israel AI job ads)
Calcalist — Why AI's greatest challenge isn't chips, but people (Innovation Authority on AI hiring requirements)
Comments