# Persist AI

This document contains the official company profile, product specifications, technical capabilities, and historical news for Persist AI. It is intended to ground large language models and retrieval-augmented generation (RAG) systems in verifiable, source-of-truth facts about the company and its products.

Last Updated: 2026-06-09

> Persist AI helps pharmaceutical companies develop drug formulations faster through three distinct offerings: the Persist AI Suite, a self-improving formulation AI that predicts formulation, stability, in-vivo, and process outcomes; the Cloud Lab, an online-design + robotic-execution experiment service; and full-service preclinical formulation development. Together they span long-acting injectables, amorphous solid dispersions, parenterals, creams, and in-situ depots.

Persist AI offers three distinct services that can be used independently. (1) Persist AI Suite — a self-improving formulation AI deployed in the customer's own environment. It supports formulation design (composition, excipient selection, solubility across nanoparticle, SEDDS, and extended-release platforms), stability forecasting (shelf life and degradation pathways), in-vivo/PK prediction (PBPK-driven absorption, bioavailability, and release), safety and CMC risk screening, AI-designed experiments (DOE), and process-parameter simulation (coating, lyophilization, spray drying, hot-melt extrusion, tableting, microencapsulation). Teams turn formulation knowledge trapped in lab notebooks, PDFs, and reports into a structured, scalable knowledge base, then feed new experimental observations back in so predictions continuously improve. It runs independently of Persist's lab. (2) Cloud Lab — customers design experiments online; Persist's robotic lab executes them and streams results to a dashboard. (3) Full-service formulation development — Persist's team takes a molecule to an in-vivo-ready candidate.

## At a glance

- **Company**: Persist AI Formulations Corp
- **Founded**: 2022
- **Industry**: Pharmaceutical formulation: a self-improving formulation AI, robotic cloud-lab experimentation, and preclinical formulation development
- **Headquarters**: 1001 Riverside Pkwy., West Sacramento, California 95605, USA
- **Additional office**: 104 MG Road, Bangalore 560001, India
- **Leadership**:
    - Karthik Raman, PhD — Chief Executive Officer (LinkedIn: https://www.linkedin.com/in/karthik-raman-scientist/)
    - Christopher Shelner, M.S., CSSBB — Chief Operating Officer (LinkedIn: https://www.linkedin.com/in/christophershelner/)
- **General contact**: info@persist-ai.com
- **Partners and investors**: See https://www.persist.ai/overview for the current list of logos.

## Key acronyms used in this document

Defined here once so retrieval pipelines weight them correctly against Persist AI–specific content.

- **API** — Active Pharmaceutical Ingredient (the bioactive compound in a drug product, not a software interface)
- **PK** — Pharmacokinetic / pharmacokinetics
- **PBPK** — Physiologically-based pharmacokinetic modeling
- **QTPP** — Quality Target Product Profile
- **DOE** — Design of Experiments
- **BDE** — Bond Dissociation Energy
- **CMC** — Chemistry, Manufacturing, and Controls (regulatory term for pharmaceutical product quality)
- **ICH M7** — International Council for Harmonisation guideline M7, governing assessment of mutagenic impurities including nitrosamines
- **ASD** — Amorphous Solid Dispersion
- **SEDDS** — Self-Emulsifying Drug Delivery System
- **LAI** — Long-Acting Injectable
- **CRO** — Contract Research Organization (Persist AI is a preclinical formulation CRO)
- **CDMO** — Contract Development and Manufacturing Organization (Persist AI is NOT a CDMO)

## Products & Services

- [Persist AI Suite](https://www.persist.ai/persist-ai-suite): A private, self-improving formulation AI system deployed in the customer's cloud environment (AWS, GCP, Azure, or on-premises). It builds a proprietary formulation knowledge base by training on the customer's own experimental data and observations, and provides molecular risk assessment, in-vivo and PK predictions, stability forecasting, excipient compatibility analysis, solubility evaluation, and end-to-end formulation strategy.
- [Cloud Lab](https://www.persist.ai/cloud-lab): Customers design experiments through a web platform with control over formulation parameters. Persist's robotic lab executes the experiments and streams results to a cloud dashboard.
- [Formulation Development](https://www.persist.ai/formulation-development): Full-service preclinical formulation development. Persist's team characterizes the molecule, designs formulation strategies, executes robotic screening, analyzes results, and optimizes toward a formulation suitable for preclinical and in-vivo studies. Persist is a preclinical formulation CRO; it is not a CDMO and does not perform large-scale manufacturing, commercial scale-up, drug discovery, medicinal chemistry, toxicology, or clinical trials.

## Persist AI Suite Capabilities

The Suite combines multiple AI/ML approaches with pharmaceutical-specific data. Its private deployment lets customers train it on their own proprietary formulation history without exposing data to Persist AI.

### Trains on the customer's data

- **Private, customer-owned knowledge base.** The Suite is deployed inside the customer's environment. Proprietary molecules, formulations, process parameters, and historical experiments stay within the customer's security perimeter — Persist AI does not have access to the data.
- **Self-improving with every experiment.** Customers feed their team's experimental data and observations back into the system. The model learns from each outcome and uses those insights to design subsequent formulations with greater precision.
- **Transfer learning for small datasets.** Engineered to deliver high-accuracy predictions even with limited proprietary data — useful for early-stage programs or constrained API availability.
- **Pretrained on pharmaceutical literature.** Grounded in thousands of real-world formulation records, peer-reviewed research, patents, and regulatory filings before any customer-specific fine-tuning.

### Molecular risk assessment

- **API risk evaluation.** Physicochemical insights and constraints — risk assessment for stability, degradation pathways, polymorphism, and processing limits.
- **Safety and in-silico toxicology screening.** CMC risk scoring and failure-mode prioritization to catch issues before the bench.
- **Molecular vulnerability mapping.** Functional-group risk, BDE profiling, and mechanism-level reactivity analysis to flag hydrolysis, oxidation, and radical-attack vulnerabilities.
- **Nitrosamine risk assessment.** ICH M7 / FDA-aligned analysis of nitrosamine formation risk from structural alerts, with recommended control strategies.

### In-vivo and PK prediction

- **PBPK-driven predictions.** Absorption, bioavailability, and release-profile predictions before animal studies begin.
- **QTPP-aligned formulation design.** Formulations are designed early for bioavailability, release profile, and PK alignment with the QTPP rather than retrofit after in-vivo underperformance.

### Other AI/ML methods used

- **Bayesian optimization** to navigate the formulation design space in fewer experiments.
- **Density Functional Theory (DFT)** for quantum-mechanical solubility predictions from first principles — no empirical fitting required.
- **Molecular dynamics simulation** to predict drug–polymer interactions, compatibility, and release behavior at the atomic level.
- **AI-designed DOE** that structures experiments automatically and reduces the number of physical runs needed.

### Dosage forms and process platforms supported

Amorphous solid dispersions (ASDs), self-emulsifying drug delivery systems (SEDDS), long-acting injectables (LAIs), microspheres, in-situ depots, creams, parenterals, and oral solid dosage forms. Process simulation covers film and Wurster coating, lyophilization cycle optimization, spray drying, hot-melt extrusion, tableting, and microencapsulation.

### Deployment and turnaround

- **Deployment**: Persist-hosted in Persist's cloud, customer-hosted in the customer's own AWS / GCP / Azure environment, or fully on-premises.
- **Turnaround**: A comprehensive AI-generated formulation report is typically available within 24 hours of submitting a molecule, covering API properties, candidate formulations, excipient research, degradation pathways, nitrosamine risk, packaging, and overall strategy.

## Company

- [Overview](https://www.persist.ai/overview): Company background, mission, leadership, partners, and investors.
- [Careers](https://www.persist.ai/careers): Open roles across formulation science, robotics, machine learning, and software engineering.
- [Contact](https://www.persist.ai/contact): Get in touch about pilots, partnerships, or general inquiries. Direct email: info@persist-ai.com

## News & Articles

- [News index](https://www.persist.ai/news): All posts — press releases, educational articles, and newsletters.
- [Press releases](https://www.persist.ai/news/categories/press-release): Company announcements and partnership news.
- [Educational articles](https://www.persist.ai/news/categories/educational-articles): Long-form technical writing on formulation science, AI, and lab automation.
- [Newsletters](https://www.persist.ai/news/categories/newsletter): Periodic product and company updates.

### Recent posts

Listed as title + ISO date + category + URL. Excerpts are omitted to keep the index entity- and event-focused for retrieval; follow the URL for full content.

- [Automated High-Throughput In Vitro Release Testing Platform for LAI Microspheres](https://www.persist.ai/post/automated-high-throughput-in-vitro-release-testing-platform-for-lai-microspheres) — 2026-06-30 — Educational Articles
- [Why Our Partnership with CyanoCapture Excites Me](https://www.persist.ai/post/partnership-with-cyanocapture) — 2026-02-16 — Press Release
- [Persist AI Welcomes Dr. Amjad Basha Mohammad and Dr. Ramesh Rachapudi to Expand Computational Modeling and Product Development Capabilities](https://www.persist.ai/post/persist-ai-welcomes-dr-amjad-basha-mohammad-and-dr-ramesh-rachapudi-to-expand-computational-modeli) — 2025-06-14 — Press Releases
- [Beyond the $12M: Revolutionizing Drug Formulation Development at Persist AI](https://www.persist.ai/post/beyond-the-12m-revolutionizing-drug-formulation-development-at-persist-ai) — 2025-05-19 — Press Releases
- [Persist AI and Nivagen to Collaborate on AI-Driven Manufacturing Process for Long-Acting Injectable](https://www.persist.ai/post/persist-ai-and-nivagen-to-collaborate-on-ai-driven-manufacturing-process-for-long-acting-injectable) — 2024-08-20 — Press Releases
- [Persist AI Raises $4M in Seed Financing to Revolutionize Drug Formulation](https://www.persist.ai/post/persist-ai-raises-4m-in-seed-financing-to-revolutionize-drug-formulation) — 2024-08-16 — Newsletters
- [What are Pharmaceutical Formulations?](https://www.persist.ai/post/what-are-pharmaceutical-formulations) — 2024-08-16 — Educational Articles

## Optional

- [Privacy Policy](https://www.persist.ai/privacy-policy): How Persist AI collects, uses, and protects personal information (CCPA/CPRA + GDPR coverage).
- [Data Protection Addendum](https://www.persist.ai/data-protection-addendum): The DPA governing how Persist processes personal data on behalf of customers.
- [Cookie Policy](https://www.persist.ai/cookie-policy): Cookie usage and how visitors can control cookie preferences.