Inovia Bio shaved more than three years off drug development with AI guardrails
Antonio Nicolae is Chief Technology Officer at Inovia Bio, which combines deep tech and drug-development expertise to help teams move programs forward. The firm has baked its technology into every workflow for five years and says some customers have cut more than three years from their development timelines by using tightly tested AI workflows.
Use AI for narrow, well-defined jobs on strict guardrails and test them relentlessly: Inovia executes over 100,000 tests daily and produces specialised ‘PhD AI employee’ models for market sizing, literature reviews and rapid C-suite updates, while keeping senior engineers responsible for systems architecture. Ignore broad, catch-all AI projects and the surrounding hype.
What’s the quick origin story of your business, and what makes what you do genuinely different from the other options your customers are looking at?
It takes over ten years on average to get a drug from the Investigational New Drug, or IND, application to marketing approval. Both Imi Faghmous and I felt that was too much time lost to fragmented evidence and slow decision-making. We knew, with the right technology, it could be done much faster without compromising safety. So we founded Inovia Bio. We work alongside customer teams to solve the strategic problems that can determine whether a development program moves forward, stalls, or misses an opportunity.
As for what makes us different from contract research organizations and consultancies: we sit right at the nexus of deep tech and drug development expertise, which is a rare combination. For the past five years, our tech has been baked into every workflow, so when we hand a customer a solution, it comes with a quality stamp. No change orders, none of those projects that somehow never end. We put the customer first and move fast. Fast enough that some of our customers have shaved more than three years off their drug development timelines with us.
Where does most of your new business come from today, and where do you wish it came from?
We consider ourselves lucky here, as most of the new business comes from "word-of-mouth" referrals. And we couldn’t be happier about it. In an industry built on trust, a referral is the truest indicator of the quality of your services. Where do I wish it came from? More of the same. There is no better compliment than a customer sending their peers to us.
What’s one thing you’ve actually handed over to AI in the last year, and one thing you tried it on and went back to doing the human way?
As a veteran of the field, I’ll start by saying "AI" is a very broad term, so I’ll assume we’re talking about large language models and agentic workflows. Picking just one thing we’ve handed over is hard. We run AI on guardrails, with over 100,000+ tests executed daily on our infrastructure to ensure quality, and each month we output multiple "PhD AI employee" type models, from market sizing to literature reviews, to the chat AI in the alpha version of our IEP platform that gives users C-suite level updates within seconds. The one thing we went back to doing the human way? As a CTO, I can tell you, AI is still not as good as our senior engineers at technical systems architecture design, so that remains firmly human.
Have you noticed a change in how customers find you or what they already know before they reach out? Is anyone arriving through AI assistants like ChatGPT yet?
We do see some inbound requests coming through AI assistants. But given the nature of our business, where you’re placing your trial design in the hands of an external party, a great deal of trust is required. Those relationships take more effort to build than those with someone who arrives as a word-of-mouth-referred partner.
If you were advising someone in your industry on all of this for the next 12 months, what would you tell them to actually do, and what would you tell them to ignore?
What to do: focus on AI workflows that achieve one job really well, put them on guardrails, and test them relentlessly.
What to ignore: the hype and the "catch-all" approaches. Hype is dangerous in our industry because patients can be the ones feeling the consequences of an uninformed decision.
Thank you to Antonio Nicolae and the team at Inovia Bio for sharing what actually worked with Leaders Perception readers.
Want to share your playbook?
Leaders Perception publishes founder and operator interviews every week. A few questions, about five minutes, no calls. It is free and everyone we accept gets published.