# Neopoly AI

- **Website:** <https://neopolyai.com>
- **Primary alias:** `neopolyai.com`

## Description

Neopoly AI develops tools for exploring chemical compounds through analogical reasoning and active learning. Its platform mines relationships from a user’s own compound library, identifying both changes already demonstrated by measured compounds and combinations supported by the data but not yet tested together. These rules are used to generate new molecular candidates from selected starting molecules, with each candidate linked to the analogy, source compound, and supporting evidence behind it.

Neopoly predicts candidate properties by transferring measured results from relevant precedent in the library, rather than relying solely on smooth interpolation. It ranks candidates not only by predicted scores but also by their potential value in guiding future experiments. Results from experiments can then be returned to the library, informing new rules and subsequent rounds of generation and prediction.

Through neoLAB, users can explore the engine with three preloaded kinase libraries or provide their own data in a CSV file. The platform is designed to make its reasoning traceable, helping researchers review, assess, and prioritize candidates for experimental testing.

## Industries

- ⛑ Health
- ⛑ Biotechnology and Pharmaceuticals _(under Health)_
- 🧪 Science and Education
- 🧪 Chemistry _(under Science and Education)_

## Logos & icons

- logo _(primary)_ — PNG — [download](https://cdn.brandfetch.io/id4U69lt45/w/440/h/90/theme/dark/logo.png?c=1bxid64Mup7aczewSAYMX&t=1792877798264)

## Colors

| Name | Hex | Theme |
| --- | --- | --- |
| Ultramarine | #110ca9 | accent |
| Ultramarine | #120cb0 | dark |
| Cornflower | #86bde5 | light |

## Fonts

- var(--font-mono) — asset _(custom)_

---

_Source: <https://brandfetch.com/neopolyai.com>_
_See [/llms.txt](/llms.txt) for a full list of agent-readable pages, and [/auth.md](/auth.md) for how agents authenticate._