Rivia raises $15M to fix the broken data infrastructure behind clinical trials

Earlybird leads round as the startup builds a reusable intelligence layer to unify fragmented trial data and enable AI-driven workflows

Rivia, the Zurich-based data engine for clinical trial intelligence, today announced a $15 million Series A led by Earlybird, with participation from Defiant and existing investors Speedinvest, Amino Collective, and Nina Capital.

Over the past three years, the company built what it calls the first reusable intelligence layer for clinical trials.

Its data engine integrates thousands of heterogeneous data files in real time, applies trial-specific scientific logic using its proprietary library of reusable configurations, and feeds harmonised data directly into operational review workflows. This enables more proactive decision-making.

On this foundation, Rivia is launching a new suite of embedded AI agents. Its first agent, Spark, instantly converts natural language into publication-grade clinical visualisations. Next-generation agents are being deployed in proactive data quality monitoring and oversight workflows, enabling earlier detection of deviations, intelligent prioritisation, and structured, auditable action.

I spoke to Erik Scalfaro, CEO and Co-Founder of Rivia to learn more.

Tightening regulation and shrinking margins

The raise comes at a critical moment for drug development. Regulatory scrutiny is intensifying, with new FDA guidance requiring clinical trial operators to manage risks and compliance proactively. The updated framework explicitly embraces innovation in trial design, conduct, and technology.

At the same time, the economics of drug development are under pressure. Industry returns have declined from 11 per cent a decade ago to roughly 3  per cent today, with the number of therapies that successfully reach the market remaining stubbornly low. Yet, the operational reality has not evolved.

Why clinical trial infrastructure is broken

Despite advances in biotech, clinical trial data infrastructure remains deeply fragmented. Many clinical trial operators continue to rely on spreadsheets and fragmented systems — a problem rooted in how the industry is structured. Clinical trial data sits across multiple vendors that lack standardised integrations. APIs are still rare, and source systems are designed for secure storage, validation, and compliance — not interoperability.

According to Scalfaro, “Teams end up downloading files from multiple systems and stitching them together in spreadsheets, or hiring programmers to build bespoke pipelines for each study.”

These pipelines can take months to build and are typically layered on generic analytics tools never designed for clinical trials.

Clinical trials have evolved — the infrastructure hasn’t

According to Scalfaro, clinical trials have changed faster than the infrastructure that runs them. Data volume has increased more than 400 per cent over the past decade. 

“Incumbent Systems like Veeva and Medidata were designed primarily to capture data for regulatory compliance, they solved the problem of their time, not to integrate and analyse data across dozens of vendors in real time.

Their architecture reflects that origin, and their business model depends on being a single central system across their product offering (Medidata has +60 products). That creates a natural disincentive to be truly vendor-agnostic or multi-source.”

However, modern trials now generate data from speciality labs, patient diaries, imaging, genomics, wearables and operational systems. 

As a result the “all-in-one” solution by a single vendor doesn’t fit those growing specialised needs.

The result, according to Scalfaro, is a fragmented stack where sponsors still stitch data together through manual patchwork.

 “Fixing that would require rebuilding the underlying infrastructure, which is why the gap between data growth and trial infrastructure has widened.” 

And then as trials become more multimodal and data-heavy, the problem is compounding — but the underlying systems haven’t evolved. “Incumbent systems were not architected or incentivised to solve this integration problem ” he shared. The result is an industry stuck between two imperfect options:

• Generic dashboards on raw data
• Custom statistical programming rebuilt for every trial

According to Scalfaro, “What has been missing is infrastructure that models the trial logic itself so data from all sources can be integrated and interpreted consistently at speed.”

That’s the gap Rivia is aiming to fill.

Why Rivia built its data engine first

AI Rivia’s approach — building a data engine before deploying AI agents — was deliberate. Scalfaro argues it’s the only way to capture the complexity and specificity of clinical trials in a meaningful way.

“It’s the only sequence that allows us to comprehensively capture the unique specifics of each trial needed to contextualise results.”

At the core is a unified data layer — the “scaffolding” — which structures fragmented inputs into a coherent system. This enables the creation of vertical workflows tailored to trial activities, before introducing AI on top.

Scalfaro explained:  “Without that structure first, AI would simply operate on poorly organised data and produce unreliable results.”

Lower costs, faster insights — and better outcomes for patients

And today, that vertical sequence of data engine-to-agents gives Rivia a structural advantage. 

“We’ve seen biotechs run global trials on Rivia and deliver measurable results, from preventing issues that would’ve cost millions to gaining earlier clarity on which patients benefit most. With every new trial, our ontology library compounds, making our system more powerful over time, shared Scalfaro.

Lower trial costs and faster insights mean therapies can reach the market sooner and more clinical programmes can be funded.

“In the near term, the direct economic benefit is captured mainly by drug developers (biotechs and pharma companies), since clinical trials are their largest cost centre.

CROs benefit as well through more efficient operations and less manual data work. In practice, it is a positive-sum outcome across the ecosystem. “

AI and the shift toward adaptive clinical trial design

Scalfaro believes that in the long run, AI could redesign how trials themselves are structured, sharing that several newer approaches, such as decentralised and adaptive trial designs, have emerged in the past few years.  

“They can bring major benefits, but only if the operational complexity can be managed efficiently, which is where AI could massively help.  

As data infrastructure improves, AI can detect patterns earlier and identify which patient groups respond best. This could lead to trials that are more targeted and adjusted as evidence emerges.“

However, he notes that clinical development requires strong scientific rigour, meaning adoption will take time as regulators build confidence.

Ultimately, Rivia’s ambition is clear: reduce clinical trial costs by up to 50 per cent by replacing manual processes with scalable agentic systems.

According to Scalfaro, a large share of cost and delay comes not from the science itself, but from the difficulty of generating and validating underlying data. Improving trial infrastructure is therefore one of the most impactful ways to accelerate medical innovation.

“At Rivia, we are building that foundation — and as demand grows, we will significantly expand the team over the next year across Zurich and Boston.”

According to Christian Nagel, Partner and Co-Founder at Earlybird, clinical trials are among the most complex and costly workflows in healthcare, yet much of the infrastructure remains fragmented and manual. 

“Rivia has built a true intelligence layer for clinical operations, unifying data and embedding agents directly into high-impact workflows.  We believe this approach has the potential to fundamentally improve trial execution, reducing costs while increasing speed and data integrity. We’re excited to support Erik and the team as they scale this new agentic foundation for global drug development.”

“When we first backed Erik and Tiago, they took on the hardest challenge first — building the data and infrastructure engine to power the world’s most complex clinical trials. They’ve delivered and more, becoming mission-critical to their customers. We’re excited to keep backing them as they layer agentic intelligence on that foundation and build the platform no clinical trial can run without,” comments Andrea Zitna, Lead Partner for Health & Bio at Speedinvest.

Lead image:  Rivia Founders Erik Scalfaro and Tiago Kieliger. Photo: uncredited.