Generate novel, synthesizable small molecules tailored to your target
Enki™ rapidly generates potent, selective, and synthesizable lead-like structures pre-trained on 760 drug targets.
Now applicable to proximity-based therapeutics and novel payloads for antibody drug conjugates.

What is Enki™?
Enki™ is a generative AI platform purpose-built for small molecule drug discovery.
Trained on millions of experimental and computational data points from 1,000+ drug targets, Enki™ supports the rapid generation of novel, potent, selective, and synthesizable lead-like structures across 760 pre-trained drug targets from key target classes, including kinases, GPCRs, proteases, oxidoreductases, hydrolases, and ion channels.
With Enki™ 4, the platform now extends its application to degraders, PROTACs, molecular glues, and novel payload design for antibody drug conjugates (ADCs) and related modalities.
How it works?
Enki™ generates novel, lead-like structures designed to meet the defined target product profile from the outset. The generated compounds are optimized across potency, selectivity, ADMET, and synthetic feasibility before entering the lab.
Rather than screening existing libraries and committing to years of iterative refinement, R&D teams start with higher-quality leads and move directly into lead optimization.
The result: fewer design–make–test–analyze (DMTA) cycles, faster timelines, and a higher probability of success.

1. Define the preclinical target product profile (TPP)
Specify the On/Off-targets and physico-chemical properties of the molecules as input for Enki™ generative AI.

2. Enki™ generates compounds
Enki™ generates novel and diverse structures that meet the defined TPP.

3. Make your selection
Pick compounds you want to synthesize and test. Enki™ can also perform hyper-efficient lead optimization while constrained to a defined scaffold.
Why Enki?
Unprecedented novelty
Go beyond the limits of screening libraries. Enki™ explores uncharted regions of chemical space, generating novel molecular structures unlikely to emerge from traditional methods.
Drug-like and synthesizable
Every molecule Enki™ generates is designed with synthetic feasibility (90% synthetic success rate) and drug-like properties in mind — producing real, actionable lead candidates, not theoretical curiosities.
Faster hit generation
Generate ~100 lead-like compounds per run in weeks, not months. These are not screening hits. They are leads ready for optimization.
Minimal data required
Works even when data is sparse, noisy, or non-existent—ideal for novel or difficult targets.
Flexible, human-in-the-loop workflow
Enki™ integrates into your process like a design-on-demand partner—allowing chemists to guide, review, and iterate on outputs.
Scaffold-constrained lead optimization
Enki™ performs hyper-efficient lead optimization while constrained to a defined scaffold—enhancing key properties without altering your core structure.
From early discovery to lead optimization
Enki™ supports drug discovery teams with lead generation and lead optimization, eliminating Hit ID and Hit-to-lead.
Whether you need novel starting points for a new program or scaffold-constrained optimization of an existing series, Enki™ adapts to where you are in the discovery process.

Use Cases

Update: Advancing CNS-penetrant ATR inhibitors with Enki™
Building on their initial hit series, Variational AI and Rakovina Therapeutics share promising new results on ATR inhibitors generated with Enki. The latest findings reveal compounds that balance potency, selectivity, stability, and early CNS exposure—showcasing… Read More about Update: Advancing CNS-penetrant ATR inhibitors with Enki™

Variational AI & Life Chemicals join forces to discover selective dual EGFR/FGFR1 inhibitors using generative AI
Following their successful 2022 collaboration, which targeted the SARS-CoV-2 main protease, Variational AI and Life Chemicals have partnered for a second time to discover de novo generated selective dual EGFR/FGFR1 inhibitors. Read More about Variational AI & Life Chemicals join forces to discover selective dual EGFR/FGFR1 inhibitors using generative AI