The Solubility Company and Persist AI Partner on Microgram-Level Drug Formulation

Two companies operating at the intersection of pharmaceutical chemistry and automation announced a strategic collaboration on May 13, 2026. The Solubility Company and Persist AI unveiled plans to integrate single particle analysis technology into an automated cloud laboratory platform. The partnership aims to address a persistent bottleneck in drug development: obtaining reliable solubility data when material is scarce.

The core problem is straightforward. Up to 90% of pipeline drugs face significant solubility challenges. Yet traditional measurement methods require sample quantities that simply do not exist in early discovery stages. Scientists often cannot test formulations because they have exhausted their precious milligrams of compound before meaningful data emerges.

The Solubility Company developed the SPA® (Single Particle Analysis) platform to measure solubility using microscopic sample quantities. The technology enables high-fidelity measurement across the entire preclinical formulation space using a single milligram of sample. By integrating this capability into Persist AI's high-throughput robotic Cloud Lab, the collaboration creates what the companies describe as a feedback engine for fast-tracking assets to IND.

According to the official announcement, the partnership will launch the first fully automated SPA® platform. This represents a shift from empirical trial-and-error to a digital-first development cycle. The ability to generate thousands of data points at the milligram scale allows for early identification of optimal delivery vehicles months earlier than previously possible.

Dr. Sami Svanbäck, CEO of The Solubility Company, emphasized the data quality imperative. "The industry is in a data renaissance, with data-hungry AI only as good as the physical evidence it is trained on." By automating the SPA® platform, the companies are building a high-throughput engine for high-fidelity data. This matters because most existing preclinical-relevant datasets are either too small, inconsistent, or lack standardized annotation necessary for robust machine learning.

Karthik Raman, CEO of Persist AI, framed the collaboration as replacing slow-moving trial and error with a high-velocity empirical augmented AI feedback loop. "By integrating automated SPA® into our Cloud Lab, we are cutting months of repetitive downstream wet-lab experimentation by enabling scientists to derisk formulation upstream when they only have 1 milligram of material." The physical reality here is tangible: researchers no longer need to wait for synthesis teams to produce more compound before testing whether a formulation will actually work.

The collaboration delivers three key outcomes. First, automated SPA® integration: Persist AI is engineering robotics to integrate the SPA® Platform into its automated Cloud Lab infrastructure. Second, fast-track to IND: global pharma and biotech clients will access the automated SPA® Platform through the Cloud Lab to rapidly advance promising assets to dosable formulations. Third, AI model co-development: both companies will use non-proprietary high-fidelity datasets to co-develop advanced predictive AI models specifically designed for novel modalities in preclinical relevant settings.

This partnership arrives at a critical inflection point for AI in drug development. While AI is transforming molecule discovery, the industry faces critical data scarcity in physical chemistry. The gap between computational predictions and experimental validation has created a bottleneck where promising molecules stall because formulation data cannot be generated quickly enough. (This has been a frustrating reality for formulation scientists for years.)

The technical integration requires significant engineering work. Persist AI's robotics must accommodate The Solubility Company's machine vision technology while maintaining the ultra-low sample consumption that makes SPA® valuable. The companies describe this as enabling high-throughput solubility screening in preclinical vehicles. In practice, this means robotic arms handling microgram quantities with precision that manual pipetting cannot match.

Partners interested in pilot programs are encouraged to contact the respective business development offices. The companies have not disclosed pricing, timelines for commercial availability, or specific client commitments. This is typical for early-stage partnerships where the technology integration itself remains unproven at scale.

The broader implication extends beyond these two companies. If the integration succeeds, it establishes a new standard for formulation machine learning. The ability to generate standardized, high-fidelity datasets at the milligram scale could reshape how pharma approaches the discovery-development interface. Other players in automated lab technology will need to respond.

Whether this partnership actually delivers on its promises remains to be seen. The pharmaceutical industry has witnessed countless automation initiatives that struggled with real-world complexity. The difference here is the specific focus on material scarcity—a genuine constraint that cannot be solved by software alone. Whether users actually pay for the service remains the real question.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn