You bring the question and the materials. We design experiments, execute on cloud-lab automation, capture multimodal readouts, and deliver mechanism. Iterative, AI-orchestrated. 4–8 weeks per program, 2–3 cycles.
Compound-specific mechanism on proprietary compounds, in iPSC-cardiomyocytes. Multimodal readouts beyond binary risk flags.
CRISPR or small-molecule libraries. You design the screen; we run the wet-lab side end-to-end. Grant-funded co-development works when your lab doesn't have the bandwidth.
AI-bio research, functional genomics, other molecular biology questions needing iterative multimodal work. Talk to us.
For pharma cardiac safety: open atlases can't include your proprietary compounds. Internal hERG and CiPA flag risk; they don't deliver mechanism. Generic CROs execute; they don't iterate. Each engagement runs in human iPSC-cardiomyocytes with multimodal mechanism profiling (Cell Painting + DRUG-seq), active learning across 2–3 cycles within a 4–8 week window. NDA-protected.
Your first engagement runs as a calibration. You send compounds with known cardiac outcomes (clinical-positive and clinical-negative), blinded to us until delivery. We return mechanism profiles plus a fine-tuned model you keep. Low commitment to find out where the engine works on your chemistry before you scale up. NDA-protected.
Each iteration compounds. The model learns which experiments tell you the most about your compound and picks the next round accordingly.
Daniel Reda. Two prior exits in life science data: CureTogether (acquired by 23andMe) and Redasoft (acquired by Hitachi). Background in Molecular Genetics.
Scientific Advisory Board forming.