Why Ramp built its own in-house coding agent, Inspect
Source
newsletter.pragmaticengineer.com
Author
Gergely Orosz
Date
Context
Ramp built Inspect around a remotely configured development environment that gives its coding AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → access to internal context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → and verification tools. In an interview with the engineering team, the company describes agents running tests, inspecting telemetry and presenting screenshots and live previews. Central setup also removes the need for each user to prepare a local environment.
This explains the investment in a custom agent: the surrounding environment supplies company-specific resources needed to investigate and check changes. Ramp reported that Inspect sessions originated 75% of merged pull requests. That is an adoption measure for a workflow; it does not mean 75% of code was produced without human involvement.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
