i2Coalition Member Spotlight: Kadoa

Adrian Krebs, Johannes Engler, and Tavis Lochhead were frustrated with the status quo of web scraping: the fragile data pipelines, the technical debt, the unreliable results. In 2003 they founded Kadoa to fundamentally change how to work with web data. We caught up with Kadoa co-founder and CEO Adrian Krebs to discuss the unique challenges of servicing the financial sector, the timeless fundamentals of ethical web scraping, and the value in actively shaping the evolution of this industry.

i2Coalition: Can you give us the elevator pitch for Kadoa?
Adrian Krebs: Kadoa is building the web data layer for finance. Our AI agents write, monitor, and maintain mission-critical web data pipelines that produce verified datasets for investors. The vision we’re working towards is a finance knowledge graph where AI agents can query any public data that matters for investment decisions. Many of the world’s leading hedge funds and asset managers are already using Kadoa.

i2Coalition: Now that almost anyone can deploy an AI agent, what’s the number-one piece of advice you’d give for pointing those agents at the Internet?
AK: Stick to the practices the web scraping industry has refined over decades. The basics are: stay on public pages, disclose who you are (user agent), respect the source (robots.txt, rate limits at or below human browsing levels), verify outputs to catch hallucinations, and log provenance so you can always answer “where did this number come from, and when?”. I’ve read plenty of horror stories about AI crawlers hammering websites and causing real costs and downtime for their owners. The web only stays open if all data collection players adhere to these rules and standards.

i2Coalition: Kadoa positions itself as “the web data layer for finance”: what motivated you to focus on that particular client-industry, and what peculiarities did you encounter in web data extraction for that audience?
AK: Investment firms have a very large and constant need for fresh, accurate web data and the lowest tolerance for it being wrong, which makes them the hardest and best customers to build for. 

In finance, if data is wrong, late, or missing somebody is going to feel it downstream, so our customers require very strict SLAs and QA. We had to invest a lot of engineering and research effort into delivering provably correct data, meaning every data point is audited against the source, runs against automated quality checks, and is reviewed by a human if the confidence is not high enough.

i2Coalition: What impact do you think recent geopolitical events have had on your more recent growth?
AK: Market volatility generally increases demand for data. For example, when tariff announcements started moving markets, our customers needed pricing and supply-chain changes within minutes of publication.

i2Coalition: Why did your team find it important to join the i2Coalition?
AK: Because most of our customers are regulated institutions, ethical collection is a hard requirement for us and not just a nice-to-have. We’d rather help shape the industry’s standards than have standards written about us by people who’ve never run a crawler. The web data industry needs a visible line between responsible collection and abuse, especially now that anyone can point an agent at the internet. i2Coalition works to keep the internet open while enabling businesses to build on public data, so joining was the logical step for us.

i2Coalition: Which i2Coalition initiatives connect most deeply with your team’s ethos?
AK: The principles of the EWDCI map almost one-to-one to what we believe in: legality and privacy by default, rate limits that work for both the collector and the website, and collecting data while considering its potential misuse. We built self-healing, validation, and compliance controls into the platform from the beginning, which aligns closely with these principles. Beyond EWDCI, the policy work around web access matters most to us: the norms being set for AI agents right now will define this industry for a decade, and the responsible-actor perspective needs to be in that room.

i2Coalition: Where does government policy interface with your work on a daily basis?
AK: At two levels. At the collection level we serve global institutions, so we constantly track evolving US, EU, and global rules on public-data access, along with the emerging norms around AI crawlers. 

At the customer level, our clients are regulated companies that rely on us as a trusted, ethical partner. Auditable provenance, compliance policy enforcement, and data-handling controls are built into every pipeline, because a compliance officer will eventually ask.