What Ramp's internal AI suite teaches us about what AI-first actually means.
AI tool adoption across Ramp. Every employee. And then they noticed something concerning.
"Most people were stuck. It wasn't that the models weren't good enough or that people lacked ambition."
Seb Goddijn, Ramp
Terminal windows, npm installs, and MCP configurations were too much for most people to navigate. The few who pushed through had wildly different setups, with no way to share what they'd learned. They'd created urgency without infrastructure.
"The models are already exceptional. Most people use them like driving a Ferrari with the handbrake on."
Seb Goddijn, Ramp
An internal AI productivity suite designed around three principles. Not to simplify, not to dumb it down. To raise the floor for everyone at once.
The default approach for non-technical users is to simplify: put the product on rails, offer fewer options, make it dummy-proof. Ramp couldn't disagree more. Power users thrive on multi-window workflows, deep integrations, scheduled automations, persistent memory, and reusable skills.
"The goal isn't to remove complexity. It's to make it invisible while preserving full capability."
The biggest failure mode wasn't that people couldn't figure things out. It was that everyone had to figure things out alone. A workflow discovered by one person didn't help anyone else. Glass needed to compound wins into organisational capability.
Becoming an effective AI user is a skill. People improve through repetition and experimentation. But the product can accelerate that curve by suggesting the right skill at the right time, and showing what "good" looks like in the moment.
"No amount of workshops can match a targeted nudge while you're already doing the work."
Glass ships auto-configured. Sign in once via Okta SSO and every Ramp tool is available immediately. Gong, Salesforce, Slack, Notion, Linear, Snowflake. When a sales rep asks Glass to pull context from a call, enrich it with CRM data, and draft a follow-up, it just works.
"This is the unsexy foundation that makes everything else possible."
Skills are markdown files that teach an agent exactly how to perform a specific task. When a CX engineer builds a Zendesk investigation workflow, they package it as a skill. The entire support team levels up overnight. The Sensei AI guide surfaces the five most relevant skills for any new user on day one, so nobody has to browse 350 options.
When users first open Glass, it builds a full memory system from every authenticated connection: the people they work with, their active projects, and references to relevant Slack channels, Notion documents, and Linear tickets. The agent enters each conversation with the context the user expects, rather than spending time searching for it.
"A synthesis and cleanup pipeline runs every 24 hours, mining previous sessions and connected tools for updates. Glass adapts to your world without you having to re-explain things every session."
Glass turns a laptop into a server. Scheduled automations post results to Slack daily. Slack-native assistants listen in channels and respond using your integrations, memory, and skills. Headless mode handles long-running tasks: kick off a job, approve requests from your phone, results are waiting when you return. It's a split-pane workspace, not a single thread. Layout persists between sessions.
"Every feature in Glass is secretly a lesson. None of this was designed as education. But when you hand someone a tool that just works, they learn by doing."
Seb Goddijn, Ramp
When a CX team lead shares a skill and sixty reps level up overnight. When a new hire's first session already knows their team, their projects, and their tools. When someone who has never opened a terminal is running scheduled automations that would have required an engineer six months ago.
"The compounding is real, and we're only at the beginning of it."
The people who got the most value weren't the ones who attended training sessions. They were the ones who installed a skill on day one and immediately got a result.
"We don't believe in lowering the ceiling. We believe in raising the floor."
mikelitman.me · hello@mikelitman.me