Funding
Funding not disclosed
Founders
Product
Problem
Researchers in manufacturing and materials science often rely on costly lab experiments or physics‑based simulations to evaluate design choices. Exploring the space of possible designs with these expensive black‑box objective functions is time‑consuming and can miss optimal solutions that lie beyond human intuition.
Solution
SigOpt offers a cloud‑based optimization platform that applies advanced mathematical techniques, including Bayesian optimization, to efficiently search expensive black‑box functions. Users submit their design parameters and objective metrics, and the service proposes new experiments or simulation settings that are most likely to improve performance. The platform iteratively updates its surrogate model with each result, focusing resources on the most promising regions of the design space. By automating this exploration, SigOpt reduces the number of required experiments, accelerates convergence to optimal designs, and enables rapid iteration in additive manufacturing, nanomanufacturing, and novel material discovery.
Target Audience
Primary users are researchers and engineers in academic labs and industrial R&D teams focused on additive manufacturing, nanomanufacturing, and new material development.
Features
- Bayesian optimization engine that balances exploration and exploitation for costly objective functions
- API and SDK integrations for common simulation tools, lab automation systems, and custom codebases
- Real‑time surrogate modeling and uncertainty quantification to guide experiment selection
- Dashboard visualizations of design space, convergence metrics, and suggested next trials
- Support for parallel evaluation of multiple candidate designs to maximize throughput