“We are looking for founders who swim in AI”: J-Ventures on navigating Israel’s startup scene

“We have too many people who have never really practiced AI. They don’t know how to ask the questions and they don’t know how to leverage it,” explained Oded Hermoni, Co-Chair at J-Ventures. “We look for the opposite - for ones who know how to move fast with different tools and tech and know how to ask the AI the right questions.”

Hermoni joined CTech for its VC AI Survey to share how the fund is navigating the startup scene in Israel and the role that artificial intelligence plays in looking for new opportunities from founders. “We are looking for ones who ‘swim’ in AI,” he added.

Oded Hermoni

You can learn more in the interview below.

Fund ID
Name and Title: Oded Hermoni
Fund Name: J-Ventures
Founding Team: Oded Hermoni, Jim Koshland
Founding Year: 2018
Investment Stage: Seed, Series A
Investment Sectors: AI, Fintech, Cyber, Healthcare

On a scale of 1 to 10, how has AI impacted your fund’s operations over the past year - specifically in terms of the day-to-day work of the fund's partners and team members?

9 - AI is not new to us. Over the past seven years, we have invested in a range of AI companies across Silicon Valley, Boston, and Israel. Our portfolio spans generative AI platforms, such as MachEye—which we backed in 2018 and was acquired last year—and agentic AI applications, including DataRobot. We also specialize in use cases in health and agriculture, working with companies like Vista AI and BeeHero. At the time, it was difficult to explain what MachEye did—"You can ask it anything about your business," we would say, and executives would look puzzled, asking, "But what do I ask it? How do I use it?" What was once obscure is now obvious.

Among our investors, we have some of the leading experts in this field—people who have led AI development or business operations at companies like Facebook, Google, Walmart, and LinkedIn, as well as serial founders who have built and sold companies in AI.

Over the past year, we have seen a dramatic increase in deal flow for AI-centric companies, both in Israel and Silicon Valley. We’re seeing not only more startups but also product pivots as companies integrate AI into their tools in new ways. Today, AI is everywhere—across industries, business models, and workflows. On one hand, it has some of the characteristics of a bubble; on the other, it is undeniably real, evolving rapidly, and transforming nearly everything we know. This creates completely new challenges for us as investors, along with new opportunities.

Importantly, AI is also reshaping how we operate as a fund. We built—and are still developing—an internal AI-based platform to improve sourcing, screening, and supporting companies. We share this platform with another fund, and it has been much more effective in finding promising founders, analyzing opportunities, and tracking progress. We view it as a neural network of sorts, where AI connects insights across our community of over 400 professionals, helping us surface relevant expertise faster during due diligence and after we invest.

In practice, this means we can actively source deals more effectively, learn faster about markets, competition, technology, patents, teams, and use cases, and leverage our LP network to support our portfolio companies. For example, we now use generative AI to quickly produce high-quality research on companies and sectors, identify competitive dynamics, and even benchmark technical capabilities.

Have you already had any significant exits from AI companies? If so, what were the key characteristics of those companies?

Last year, we sold Macheye, which was a generative AI company and was one of the first gen AI companies back in 2018. It wasn’t an Israeli startup, though.

Is identifying promising AI startups different from evaluating companies in your more traditional investment domains? If so, how does that difference manifest?

When evaluating AI startups, we have become more selective. While the space is crowded and there is a lot of noise, we focus on companies with clear differentiation and proprietary data or technology that create a defensible moat. Everyone can fine-tune a model, but few have access to unique datasets or domain knowledge that compounds over time. In our view, that’s the real source of defensibility—not simply plugging into GPT-4 before someone else.

That’s why we’ve backed companies like Nimble Way, which is building a powerful proxy and data infrastructure layer critical for AI-driven intelligence gathering, or Vista AI, which is redefining how MRIs are interpreted with deep clinical datasets. Apera.ai applies AI to real-time automation in manufacturing, and Atidot AI brings predictive insights to the insurance industry using proprietary actuarial and behavioral data. Each of these companies represents the kind of differentiated, domain-driven opportunity we believe can withstand commoditization.

We look for three things:

• Is the data exclusive or hard to replicate?
• Does it improve model performance as the company scales?
• Does it create increasing advantage over time?

More broadly, we are seeing a fascinating shift. Some AI companies are achieving massive revenue growth, extraordinary profit margins, and customer demand at levels that would have been unimaginable even a few years ago. It’s not uncommon to see teams of fewer than 50—or even fewer than 10—generate significant profits. But with this success comes new tensions, especially around capital strategy. Often, these founders want to distribute profits, but investors prefer to keep capital inside the business to fuel growth and preserve liquidation preferences.

We encourage founders to think carefully about these dynamics and, where possible, structure their financing so profitability can reward everyone, not just fund the next fundraising cycle.

Ultimately, the best AI companies don’t just use data—they are built around it. They create AI-native experiences that feel magical, compressing ten tools into one intelligent layer. Whether in B2B or consumer applications, this isn’t just about new technology; it’s a new interface for how people experience and interact with software.

So while we see a lot of hype, our focus remains consistent: we look for real differentiation, proprietary advantage, and the capacity to scale sustainably. AI is transforming our portfolio, our investment process, and in many ways, our entire industry—and we see that as an exciting opportunity to rethink how venture capital can work.

What specific financial performance indicators (KPIs) do you examine when assessing a potential AI company? Are there any AI-specific metrics you consider particularly important?

We look at many of the same financial KPIs we’d use in evaluating SaaS companies—revenue growth, gross margins, customer retention, and sales efficiency. However, when it comes to AI companies, we go deeper into the go-to-market (GTM) motion, the differentiation of the product, and the underlying tech infrastructure.

We ask:

• Is the product sticky or just novelty-driven?
• Does the tech include proprietary data or IP that can scale and compound?
• Are the compute and infrastructure costs sustainable over time?

In short, we care not only about financial performance but also about whether the AI delivers real, defensible value and if the GTM strategy can effectively cut through the noise in a crowded space.

How do you approach the valuation of early-stage AI startups, which often lack significant revenues but possess strong technological potential?

AI can act as a real growth accelerator—some startups go from slow, single-digit growth to rapid double-digit traction almost overnight. But there’s also a lot of noise, and that hype is clearly reflected in inflated valuations.

At J-Ventures, we stay disciplined. We connect valuations to realistic exit scenarios, not just excitement or buzz. We look closely at whether the technology has long-term defensibility, and whether the team can build a real business around it, not just a great demo. If we don’t see a path to meaningful scale and exit potential, we’re cautious on price, even if the tech is impressive.

What financial risks do you associate with investing in AI companies, beyond the usual technological risks?

One of the biggest risks we see is companies relying too heavily on a use case without real proprietary knowledge, tech, or data behind it. Without defensibility, strong initial traction can quickly fade once competitors enter.

Another key risk is that AI drastically shortens development cycles and go-to-market timelines. When we see a smart product idea, we assume others are already working on something similar—it often becomes a race to market. That’s why we focus on companies with clear, long-term differentiation, and not just those who happened to ship first.

We also stay mindful of infrastructure costs, third-party dependencies (e.g., reliance on foundation models), and evolving regulatory frameworks that may affect data usage or model deployment.

Do you focus on particular subdomains within AI?

We started investing in AI in 2018, before it was a buzzword, and across different industries. From security to defence, from manufacturing to supply chain, to fintech and healthcare.

How do you view AI’s impact on traditional industries? Are there specific AI technologies you believe will be especially transformative in certain sectors?

AI is impacting nearly every traditional industry—from software development and testing to diagnostics, collaboration tools, CRM, and more. It’s like the original vision people had for 3D printing—that anyone could create anything they needed at home—but this time, AI is actually delivering on that promise.

We’re seeing both individuals and companies develop custom solutions on their own, without relying on multiple external vendors. For example, one founder we know moved all of his Salesforce data into an AI-based CRM he built himself—and completely stopped using Salesforce.

This kind of flexibility and power is transformative. AI enables not just automation, but actual replacement of traditional tools across sectors like health, enterprise SaaS, logistics, and beyond. It’s not just a tech layer—it’s a new way of building, running, and scaling businesses.

What specific AI trends in Israel do you see as having strong exit potential in the next five years? Are there niches where you believe Israeli startups particularly excel?

While Israel came a bit late to foundational AI compared to its early lead in cyber, it’s catching up fast, especially in areas where we’ve always been strong. We see significant exit potential in AI for cybersecurity, cloud infrastructure, AI-enabled developer tools, and media applications.

Israeli startups also excel when combining hardware and software, which positions them well for AI infrastructure and edge computing opportunities. Companies like Bria AI (media), Descope (cyber identity), and Nimble Way (infrastructure and data pipelines) show how Israeli tech can lead when it builds around deep domain expertise and proprietary layers.

We also see strength in AI for vertical markets—like health, agtech, and defense—where Israeli founders have both the vision and the technical depth to build globally relevant solutions.

Are there gaps or missing segments in the Israeli AI landscape that you’ve identified? What types of AI founders are you especially looking to back right now in Israel?

We have too many people who have never really practiced AI. They don’t know how to ask the questions and they don’t know how to leverage it.

We look for the opposite - for ones who know how to move fast with different tools and tech and know how to ask the AI the right questions. It can create the lean startup, the bottom-up models.

We are looking for ones who “swim” in AI.