Home›Companies›Persist AI

Persist AI

Developing long-lasting drug formulations 50% faster

It takes 5 years for pharma to develop long lasting drug injections for chronic diseases like cancer and diabetes. Persist uses AI-driven automation to reduce formulation development time down to 2 years, a ~50% reduction.
Persist AI
Founded:2022
Team Size:6
Location:Woodland, CA
Group Partner:Surbhi Sarna

Active Founders

Karthik Raman, Founder

When mom got brain cancer, I used to take her for her "infusions". For 4 hours, we would sit there while drug was being infused into her veins and she had horrible nausea and hair loss. I built Persist to eradicate this horrible experience for future patients. We package drugs so that patients take 1 injection per month or year, without an infusion. This lowers the overall dose the patient receives (reducing side-effects), improves compliance, and quality of life.
Karthik Raman
Karthik Raman
Persist AI

Christopher Shelner, Founder

Career program manager, first time founder. Focus has been in electronics, automation, and manufacturing. Super excited to be part of the YC batch!
Christopher Shelner
Christopher Shelner
Persist AI

Company Launches

TLDR

Persist helps pharma make drug injections that last up to 1 year.

Using robotics & AI, we reduce a 5-year drug formulation process to ~2 years.

Contact Us (founders@persist-ai.com) for a case-study formulation

To make injections we encapsulate drugs inside injectable microspheres made with FDA-approved materials.

Left: Image of a ~37 um microsphere made by our robotics

Right: A Schematic showing how encapsulated drugs are released over 1 year so patients don’t take as many injections.

🤔THE PROBLEM

For chronic diseases, most patients prefer 1 injection/yr rather than multiple injections per week.

However, making a long-lasting injection takes a lot of trial and error and the time between iterations is long.

Key challenges include:

  • Using manual work & intuition is slow: Typical teams manually mix and match chemicals to optimize the amount of drug that’s loaded into the spheres and the rate of release of drug
  • Trial and error is cumbersome: Because changing one variable may have multiple effects
  • Scale-up is unpredictable: Even if something works at a small scale, batch-to-batch consistency during scale-up is uncertain because production is done in large reactors

đź’ˇOUR SOLUTION

We utilize custom robotics and AI to solve the problems with building microspheres.

  • We automatically mix & optimize formulations: We use a closed-loop robotics + AI approach so there’s minimal manual work
  • We reduce trial & error with AI predictions: We generate 50x more data than most formulation teams and use that data to train AI. This unique data enables us to make better predictions than others.
  • We build small batches, faster: So we can QC batches more frequently and scale up predictably

❤️OUR ASKS

Introduce us to friends who work in Pharma. Email founders@persist-ai.com

“Follow” us on LinkedIn

Selected answers from Persist AI's original YC application for the W23 Batch

Describe what your company does in 50 characters or less.

Help pharma develop slow release drugs faster

What is your company going to make? Please describe your product and what it does or will do.

For a pharma company to develop a sustained-release (slow release) drug formulation takes about 3-5 years and $10M or more. This is because there are a lot of variables and scientists' intuition (aka guesswork) needed to arrive at the final formulation solution.

We will provide a service to pharma companies to develop their formulations for them 50% faster and cheaper than current methods. To do this we will screen formulations in high throughput using automation and train ML which can make better predictions based on data (rather than guesses).

This will enable pharma companies to get drugs to market faster and earn revenue sooner.

YC W23 Application Video

YC W23 Demo Day Video