Bystreet (YC W25) gives hotel brands, operators, and vendors the ownership data they need to reach the people behind a property. Its AI data platform connects hotels with their owners and key decision-makers, turning a fragmented research problem into structured intelligence.
For co-founder Cameron Hake, that work created two related infrastructure needs: scraping hotel websites at scale and giving Bystreet's agents reliable web data while they research the ownership structure of each property.
"We found Context.dev through the YC launch. We were looking into scraping solutions both for mass scraping and as a tool for our agents, and this fit the bill."
One scraping layer for two workloads
Bystreet uses Context.dev directly in its hotel data pipeline. The same service also supports the agents that investigate who owns each hotel, so the team does not have to maintain separate web-access systems for batch collection and agent research.
That matters because hotel ownership is rarely presented in one clean source. An agent may need to work across hotel websites and the pages around them before it can connect a property to the right owner. Context.dev gives that workflow dependable access to usable web content while also handling Bystreet's higher-volume scraping.
"We have been using it to scrape hotels and then also as a tool for our agent that goes and researches the ownership structure of each hotel."
Setup delegated to Codex
Cameron did not turn the integration into a separate engineering project. He gave the setup task to Codex, and the coding agent connected Context.dev without requiring him to troubleshoot the implementation.
"Setup was very straightforward. I just asked Codex to set it up; it had no issues. Everything just works."
That agent-led setup let Bystreet move straight from evaluating the service to using it in the workflows that matter, with no lengthy onboarding cycle in between.
Well over one million scrapes without an issue
The integration stayed straightforward after launch. Bystreet has now run well over one million scrapes through Context.dev across its hotel data and ownership-research workflows.
"It's been very plug and play. We've done well over a million scrapes with it at this point and haven't had any issues."
At that scale, reliability is not a background detail. Every failed page can leave a gap in a hotel's record or interrupt the agent trying to establish who owns it. More than one million successful scrapes gives Bystreet a web data layer it can treat as infrastructure instead of another system to babysit.
Key benefits
- Two workloads, one service: Context.dev supports both mass hotel scraping and agent-led ownership research.
- Agent-native setup: Cameron handed the integration to Codex and got a working implementation without troubleshooting it himself.
- More than one million scrapes: Bystreet has taken the integration well beyond a small trial or proof of concept.
- Plug-and-play reliability: the team has completed that volume without running into issues.
The outcome
Bystreet found a scraping service through the YC launch and put it to work across both sides of its data operation. Context.dev now helps collect hotel information at scale and gives Bystreet's agents the web access they need to research ownership structures, with well over one million scrapes completed and no issues along the way.
Building an agent or data pipeline that needs reliable access to the web? Context.dev's Markdown API turns pages into clean, usable content without making your team operate the scraping infrastructure behind it.
P.S. Need to identify the owners and decision-makers behind hotels? Explore Bystreet.