$200 GPU credits for yctierlist foundersClaim yours →
Harbor

Harbor

AI system-of-record for clinical trial data

Spring 2026ActiveHealthcareSaaSHealth TechB2BHealthcareSan Diego, CA, USA
Harbor is a new, AI-enabled system of record for clinical trial data that replaces legacy electronic data capture (EDC) systems. Today, medical device and pharma companies spend months setting up custom EDCs that research sites then spend hours per subject per day entering data into. With Harbor, the EDC is spun up simply by reading your clinical trial protocol and research sites upload raw source documents into the platform, skipping the tedious and error-prone data entry process altogether. EDCs haven’t fundamentally changed since the 1990s. They are passive, regulatory compliant data bins that cause massive trial delays, coordinator burnout, and weeks of back-and-forth to resolve data errors. Because of this, data management currently consumes ~30% of total clinical trial budgets. With LLM and document extraction capabilities finally reaching the accuracy, reliability, and traceability required in high-risk domains, Harbor is redefining the way data is collected and monitored in clinical trials, all while remaining fully GCP- and 21 CFR Part 11 compliant. After months of building our initial product, our platform is now live with its first few clinical studies. For our largest customer and their 1600 subject randomized trial, we were able to use our internal tools to get their study database configured in just one week, which was ten times faster than legacy EDC vendors quoted. We built Harbor because we lived through this problem. Albert spent four years in clinical trials and regulatory strategy for a medical device startup, spending weeks at clinical sites watching coordinators battle legacy software firsthand. Nate brings the technical firepower, having built scalable software at Google and Ramp and as employee #1 at a previous YC startup.

Note: This is a preliminary assessment based on limited publicly available information. We did not have access to LinkedIn profiles or live product screenshots for this analysis. We will update this entry with a more thorough review soon.

Verdict

High Signal
Market Opportunity
Clinical trial data management is a massive B2B enterprise market — EDC is a multi-billion dollar segment within the broader ~$50B+ clinical trials services market. ICP is clear (medical device and pharma companies running regulated trials). Data management consumes ~30% of trial budgets, meaning displacement opportunity is enormous with strong willingness to pay.
High Signal
Founder Signal
Albert has direct domain expertise: 4 years at Biolinq running clinical trials and regulatory strategy, taking a medical device through FDA De Novo authorization — exactly the problem Harbor solves. Nate has 5 years of technical experience at Google, Ramp, and as employee #1 at a previous YC startup, plus blockchain infrastructure work. This is a textbook domain + technical co-founder pairing with no gaps.
Medium Signal
Competition
Legacy EDC incumbents (Medidata Rave, Oracle Clinical, Veeva Vault EDC) dominate but haven't fundamentally changed since the 1990s — a classic innovator's dilemma setup. Harbor's differentiation is AI-native document extraction vs. manual data entry. However, these incumbents have deep regulatory compliance track records and sticky enterprise contracts, making displacement slow and sales cycles long.
Medium Signal
Product
Platform is live with first few clinical studies, which is real traction. Largest customer has a 1600-subject randomized trial, and Harbor configured the study database 10x faster than legacy EDC vendors quoted. However, no named customer logos, pricing page, or revenue metrics disclosed — 'first few' studies is early but promising.
OverallA Tier

Harbor is a strong setup: a domain expert who lived the pain (Albert's 4 years at Biolinq, including FDA authorization experience) paired with a technical founder who's shipped at scale (Google, Ramp, YC). The product is already live in real regulated studies, and the 10x configuration speed improvement on a 1600-subject trial is a compelling proof point. The market is large, B2B, and deeply underserved by aging incumbents. The main risks are long enterprise sales cycles in heavily regulated healthcare, the challenge of displacing entrenched EDC vendors with compliance-conservative pharma and medtech customers, and the gap between 'first few studies' and meaningful ARR. If they can convert that large trial into a reference customer and navigate GCP/21 CFR Part 11 validation requirements at scale, this could be a genuinely important company.

Active Founders

Albert Cai
Albert Cai
Founder

Albert is the co-founder and CEO of Harbor. Before Harbor, Albert helped run clinical trials and regulatory strategy at Biolinq, helping take the company's first and flagship product from prototype through its first human studies and eventual FDA De Novo authorization. Prior to Biolinq, Albert led diabetes technology coverage at Close Concerns, a boutique consulting firm. Albert studied biomedical engineering at the University of Michigan.

Nathan Leung
Nathan Leung
Founder

Nate is the co-founder and CTO of Harbor. He spent the last five years working at tech companies ranging from seed-stage startups to Ramp and Google. He’s built zero-to-one systems as the first employee of another YC startup and wrote smart contracts securing millions in assets at a blockchain infrastructure company. He studied pure math at UCLA and computer science at the University of Michigan.

Harbor
Harbor
TierA Tier
BatchSpring 2026
Team Size2
StatusActive
LocationSan Diego, CA, USA
$200 GPU creditsClaim yours →