Excelsior Sciences is a New York-based AI chemistry company rethinking how small molecules are discovered and made. Its core 'smart bloccs' technology, automated synthesis-friendly chemical building blocks, acts as a modular chemical language that enables rapid, machine-runnable carbon-carbon bond formation. By making synthesis programmable and automatable, Excelsior closes the loop between AI-driven design and physical chemistry, so machines can produce molecules and AI can learn from the resulting data. The smart-blocc insights originated in Marty Burke's lab at the University of Illinois. The platform targets both faster small-molecule drug discovery and more efficient, sustainable manufacturing. In December 2025 Excelsior raised $95M total, a $70M Series A co-led by Deerfield, Khosla Ventures, and Sofinnova Partners with participation from Eli Lilly and MIT, plus a $25M grant from New York's Empire State Development.