How Ramp Uses Claude

How our LLM gateway uses Thompson sampling over lognormal latency posteriors to pick the fastest, most reliable model route per request

Agents acting on behalf of a user provides a natural starting point for finance teams beginning to adopt agents, as well as for fully autonomous companies.

Marketing to agents is becoming as important as marketing to humans. We ran an experiment embedding tracked incentives on ramp.com pages addressed to AI agents. Here's what we learned.

How and why we built a real-time AI spend and usage pipeline for finance teams.

How we compare the performance of large language models on real financial tasks at Ramp, and why benchmarks built for production use cases matter.

This article explores the rise of Forward Deployed Engineering (FDE) as a strategic role in B2B tech companies, tracing its origins from Palantir to its current adoption across companies like OpenA...

The article discusses the development of Ramp's MCP server, which utilizes the Model Context Protocol (MCP) to enable natural language interaction with business data via a SQL interface.