AI is moving from lab experiments to life-saving outcomes—and Persistent Systems wants to accelerate that shift.

The digital engineering firm has announced a new collaboration with NVIDIA to bring AI-powered drug discovery into real-world production for the healthcare and life sciences (HLS) sector. The partnership focuses on applying generative AI, simulation, and agentic workflows to speed up research cycles that traditionally take months—or years.

From Wet Labs to Digital Labs

Drug discovery has long been constrained by time, cost, and complexity. Traditional R&D relies heavily on physical (wet lab) experimentation, which is resource-intensive and slow to iterate.

Persistent’s approach flips that model.

By combining its domain expertise with NVIDIA’s full-stack AI platform, the company aims to simulate biological and chemical interactions digitally—before they’re tested in the lab. That includes high-fidelity molecular modeling and large-scale virtual screening, allowing researchers to evaluate thousands of potential compounds in a fraction of the time.

The goal isn’t to replace lab work—but to make it smarter, faster, and more targeted.

Agentic AI Enters Drug Discovery

At the center of this push is Persistent’s new solution: Generative Molecules and Virtual Screening (GenMolVS).

Built on NVIDIA BioNeMo and the NVIDIA NeMo Agent Toolkit, GenMolVS uses domain-specific AI models to simulate molecular properties and generate new compounds. But the more interesting layer is what Persistent calls “agentic workflows.”

These AI agents don’t just generate data—they actively participate in the research process, continuously making decisions across stages like:

• Virtual screening of compounds

• Candidate prioritization

• Experimental planning

This creates a closed-loop system where AI models refine hypotheses in real time, helping researchers move from simulation to actionable lab experiments faster.

In practical terms, that could compress early-stage discovery timelines from months to days.

Infrastructure Built for Regulated AI

Healthcare AI isn’t just about performance—it’s about compliance, traceability, and reliability.

To support production-grade deployments, Persistent is tapping into NVIDIA’s enterprise stack, including AI Enterprise software, accelerated compute, and NIM microservices. The infrastructure is designed to handle large-scale simulations while meeting the strict regulatory requirements of life sciences environments.

The company also plans to integrate NVIDIA Nemotron models to further enhance simulation accuracy and scalability.

That combination—AI models, infrastructure, and governance—is critical for moving beyond proof-of-concepts into regulated, mission-critical workflows.

Industry Context: AI’s Growing Role in Pharma R&D

Persistent and NVIDIA aren’t alone in targeting this space.

Pharma giants and tech players alike are investing heavily in AI-driven drug discovery, with platforms from companies like Google DeepMind and Microsoft pushing advances in protein modeling, genomics, and clinical research.

What sets this collaboration apart is its focus on operationalizing these capabilities—bringing them into enterprise workflows rather than keeping them in research silos.

That’s a key shift. As the industry matures, the competitive edge will come not just from better models, but from the ability to integrate AI into end-to-end R&D pipelines.

Why This Matters Now

The pressure on healthcare and life sciences organizations is intensifying. They’re expected to deliver new therapies faster, reduce costs, and navigate increasingly complex regulatory landscapes—all while dealing with massive datasets.

AI offers a way forward—but only if it can scale.

By focusing on production-grade systems, Persistent and NVIDIA are targeting a critical gap: turning promising AI experiments into reliable, repeatable processes that can support real-world drug development.

Beyond Technology: Building AI Talent

The partnership also includes a talent component, with Persistent planning to expand its AI and LLM engineering capabilities through NVIDIA’s training and certification programs.

That’s a strategic move. As demand for AI in life sciences grows, the shortage of skilled practitioners could become as much of a bottleneck as the technology itself.

Bottom Line

AI-driven drug discovery has been a promise for years. What’s changing now is the push toward making it operational.

Persistent and NVIDIA’s collaboration signals a broader industry transition—from experimental AI models to production-ready systems that can meaningfully impact how therapies are discovered.

If successful, that shift won’t just speed up research—it could reshape the economics and timelines of bringing new drugs to market.

Get in touch with our MarTech Experts.