Antler backs healthtech founders at the beginning of their journeys. Sarah Finegan and Bazil Azmil explain how they spot talent and separate real-world utility from mere novelty.
Antler is one of the only investors in the room that backs founders before they have a product, co-founder, or even a fully formed idea. That gives the early-stage venture capital firm a early read on where the next generation of European health innovation is coming from, and who is building it.
The firm’s current British portfolio reads like a who’s who from the future and includes Prima Mente, an AI biology company building multimodal foundation models for neurological disease; healthtech firm Synthax; Bleep64, a gamified platform simulating patient and ward interactions; home care agency Gladys; Klaris, an AI-native platform to automate regulatory compliance and technical documentation for medical device manufacturers; and biotechnological research company CipherX Technologies.
Here, Sarah Finegan, associate partner at Antler, and Bazil Azmil, principal at Antler, talk to Healthcare Today about how they evaluate founders, what entrepreneurs who come from a bureaucratic environment need to unlearn, and how they evaluate novelty versus real-world use.
Antler backs founders before they even have a co-founder or a fully formed idea. What does day zero look like in practice? What are you evaluating?
Sarah Finegan: At the very beginning of our journey with healthtech innovators, we over-index heavily on founder talent. When we meet entrepreneurs looking to build the next big thing, we are looking specifically for founder DNA.
When we manually screen applications, we look closely at what an individual has achieved in their lifetime. What is their founder’s spike, their internal superpower? Is it deep domain expertise, a commercial skill set, or an exceptional technological talent?
Those who spark our excitement are invited to a first interview. In that initial conversation, we drill down into what defines them as a founder.
We also look for first-principles thinking: can they break down problems to an extent that makes them uniquely placed to solve them? If we build conviction that an individual scores highly across these core DNA characteristics, we invite them to a partner interview.
During this stage, we evaluate their broader thinking around execution. What kind of co-founder do they need to balance their skillset? Who is that perfect partner to join them on the journey and help execute the vision? We look at how they approach the zero-to-one phase, but we also look far beyond that. How do they plan to go to market? We have to consider commercial velocity: can this business scale to £100 million in five years?
Once both the investment team and the partners have built conviction, we invite the founder to work closely with us in our residency. This is a curated group of individual founders brought together in our offices, where we work side-by-side with them in the trenches, acting as sparring partners to help them get further, faster.
We then bring the partners together and, in all cases, invite a relevant external VC to join the discussion. The founders pitch for 12 minutes, followed by a Q&A session.
“The first question is always: Is it big enough to be venture-scale?”
What are the non-negotiables that make you say yes to a founder at that stage?
Bazil Azmil: In terms of non-negotiables, the first question is always: Is it big enough to be venture-scale? We have to look at it and ask, can this really scale to a £100 million annualised business? Alongside that, we need to understand if this is truly the team to win in that space.
First of all, the massive scale needs to be possible. Secondly, this team needs to be the potential winner, and the framework we look at is quite specific. Imagine if we were to see a hundred competitive teams all chasing this one exciting, large problem area. Can we look at this particular founding team and genuinely see them as the potential winner? What that actually ends up looking like in practice will differ depending on the business, the problem area, and the potential solution. You have to figure out what is going to be the most difficult part of building this business, and then look at whether this founding team has the existing knowledge, the existing network, or the specific skill set to overcome it.
When evaluating a day zero founder who is a brilliant clinician or researcher but lacks a commercial background, how do you evaluate their capacity to scale into a venture-backed CEO?
Bazil Azmil: We get the chance to observe start-ups, work with them and spar with them in those four to eight weeks before we head to an investment committee. This means we can actually see how they are adapting and how they are learning to sell.
What we look for, again, is their rate of learning and their ability to build a track record in real time. They may not have had an opportunity in the past to handle sales or commercial work, but in this incredibly short period of time, they are going to do some commercial work. For example, can they get in front of tens or hundreds of potential customers in an unbelievably short space of time? Can they figure out how to find them and how to book time with them? From there, can they start to have meaningful conversations and build a pipeline that turns into early design partners or early customers?
Having the ability to work so closely with them helps us assess their likelihood of becoming a highly commercially oriented founder.

In companies like Synthax or Klaris, the founders come from bureaucratic environments. What is the biggest cultural mindset shift these founders have to unlearn?
Bazil Azmil: Many founders are coming from regulated environments, and that often comes at the cost of velocity. In a way, that mindset is still important and good to have if you are building in the healthcare technology space broadly speaking. There is simply a higher bar for these businesses when it comes to data privacy, data security and regulatory compliance, so keeping that perspective is essential.
Where the shift really needs to happen is in working out where the parts of the business are where you have more creative licence.
Another major shift applies to people coming from these heavily regulated backgrounds who are used to a waterfall approach to building. Because they are accustomed to understanding end-to-end product development, there is a tendency to build under wraps and only launch with a fully-fledged product. However founders need to pivot and build in a highly iterative way – prototyping and securing customer feedback as much as possible along the way.
This is especially true for people building AI products today. When it comes to the ultimate integration and end goal, not everyone has a fully clear picture yet of what the final iteration looks like. Because of that, it is important to stay close to your customer and gather feedback throughout the process, ensuring you are building towards a joint, shared vision of what the solution should be.
Deep tech and frontier biology usually require capital and years of r&d before hitting clinical trials. How do you balance that with milestones required for seed or Series A? There’s a tension there, isn’t there?
Bazil Azmil: As a fund, we are probably not going to do something that fits the profile of a traditional therapeutic. While we do invest in some hardware and work with deep tech, what is ultimately important to us is that there is a strong degree of scalability and a robust software component, broadly speaking. Within that scope, we are highly sector-agnostic and incredibly open-minded.
That being said, when we work with these deeper tech companies and think about where that first check goes, what we ultimately need to figure out is how to measure progress. In cases where the technologies are highly differentiated and complex, we don’t necessarily expect immediate revenues after the initial check is invested. Instead, we ask ourselves: can this team reach milestones that we know are meaningful enough to help the business make progress towards its next funding round?
For a deeper tech business, those milestones might mean securing key partnerships, assembling longitudinal datasets to train a model, or bringing in critical hires to de-risk whatever major challenges lie downstream. Moving the ball forward in these ways still represents tangible progress, even if it isn’t immediate commercialisation of the product.
“A product needs to be more than just different; it needs to give you a reason to win.”
When backing an AI-first health company, how much are you concerned about ethical and clinical guardrails? Or does that come later?
Bazil Azmil: It is really important. Generally speaking, when we work with clinical founders, they already have a high bar for ethics. They possess an exceptionally good understanding of what should and shouldn’t be done, especially when deploying technology within the healthcare space. In terms of clinical guardrails, things like clinical robustness are viewed as a feature rather than a constraint. It is the very reason your product is good and trusted, rather than an obstacle to overcome. It is something that really gets front-loaded as much as possible.
When the entrepreneurs personally understand the problem they are trying to solve, you see that clinical robustness and clinical guidelines truly become a core feature. The risk of getting it wrong is so high, and because they feel that weight themselves, they are deeply incentivised to ensure they build the best possible product.
How do you evaluate novelty versus real-world use?
Sarah Finegan: A product needs to be more than just different; it needs to give you a reason to win. When we look at uniqueness, differentiation or novelty in the market, we have to ask ourselves: Is that difference actually going to change customers’ minds? Is it going to help close more deals? Is it going to build a more robust, larger business in the long term, or is it just different? That is a vital distinction to make. Ultimately, we need to ensure the market is sizeable and that it is a venture-scale business.



