Choose the computational discovery task your team needs: candidate design, physics-grounded scoring, reconstruction, materials data or optimization research.
Generating candidates, evaluating them and deciding what to try next are different jobs. A strong computational score is not the same as an experimentally established result.
Start with the relevant existing application. SpinForge Bio and AMDE share Discovery Core; the synthesis family supplies distinct learning, scoring, reconstruction and improvement products. Combined workflows receive their own integration and evaluation scope.
SpinForge Bio takes a supported protein target and produces the owner-described panel of 96–384 ranked candidates for downstream synthesis and evaluation. Discovery Core also supports AMDE’s distinct physics-scoring role.
Your supported target, model or input data, design constraints, intended output and the numerical or experimental criteria you will use to evaluate it.
WHAT WE EVALUATEConformance to the agreed input and output, computational quality, cost of generation and evaluation, and the domain-specific experimental or numerical checks.
Computational candidates are not evidence of experimental binding, clinical suitability or laboratory validation. Existing synthesis products remain independently available; the family architecture is not a blanket claim of measured gains on every task.
Choose an evaluation or a qualified production workflow. Compatible methods and shared checks inside its activation scope are included once. Infrastructure is separate.
See plans and deployment scopeAgree the baseline, deliverables, effort and acceptance before implementation. Request managed operation only when it is part of the desired scope.
Discuss an engagementThese are relevant capabilities to evaluate—not mandatory purchases or an assertion that every combination is already integrated.
Tell us what needs to improve, what you use today and which requirements cannot change.