HOW WE INVEST
Most investment decisions in life sciences happen at arm's length from the biology. Ours do not. Our partners have run discovery programs, read clinical data, and made go/no-go decisions on experimental platforms. We bring that operational experience to every evaluation we undertake.
We distinguish between platforms and products. A platform generates discovery repeatedly — it is a biological factory, not a single drug. We assess whether the computational or genomic engine at the center of a company can produce a portfolio of targets, not just one promising lead. Mechanism depth, generalizability, and data flywheel are the questions we ask first.
Computational biology has a reproducibility problem. We evaluate training data quality, model architecture choices, validation protocols, and whether the computational claims have been stress-tested against held-out data or orthogonal wet-lab experiments. Our partners have built and criticized these models from the inside.
Scientific novelty is necessary but not sufficient. We assess whether a discovery-stage platform has a credible path to the first meaningful clinical signal — what is the target product profile, what IND-enabling work is required, and what are the key biological questions that must be answered before the platform's promise becomes investable at scale.
Fund Strategy
Galdera has operated two funds since 2017, progressing from a $45 million debut vehicle focused on seed-stage computational biology to a $95 million second fund targeting Series A investments in AI-driven drug discovery and genomic engineering platforms. Our check sizes range from $6 million at seed to $30 million at Series A. We invest in companies before they are legible to most life sciences funds — when the platform exists but the clinical asset count is still zero.
The founders must understand the biology before the business. We are not interested in domain-agnostic entrepreneurs who hired a Chief Scientific Officer. Discovery-stage companies need scientists who can lead the science, not manage it.
Single-asset companies in early-stage biology carry enormous risk. We prefer companies where the computational or biological platform could yield multiple shots on goal — increasing the probability that at least one clinical asset advances.
Computational claims must be stress-tested. We look for companies that have validated their computational models with orthogonal experimental data — where the in silico predictions have been confirmed in wet-lab experiments and the limits of the model are understood.
We invest pre-IND, but we require that the founding team can articulate what stands between the current scientific state and a first-in-human study. Regulatory naivety at the founding stage is a significant risk factor we screen against.
We accept pitch decks and introductory conversations year-round. We respond to every company that falls within our thesis.
Connect with us