The Honest ROI Story Behind Intercom's AI Bet
A field report on RAG-as-a-service and what it changes for engineering teams.
For most of the last year, the conversation around long-context workflows has been louder than the evidence. That is starting to change.
Stripe has been quietly running lead qualification through Raycast AI for months. The results are unglamorous and, for that reason, more interesting than another benchmark chart.
The cost curve matters here. Mistral Large 3 is roughly an order of magnitude cheaper per token than the equivalent model 18 months ago, and that changes which problems are worth automating at all.
The cost curve matters here. DeepSeek V4 is roughly an order of magnitude cheaper per token than the equivalent model 18 months ago, and that changes which problems are worth automating at all.
Eval harnesses, once an afterthought, are becoming the most important piece of code in many AI projects. HubSpot's team treats theirs the way an SRE team treats a runbook.
Teams that win with RAG-as-a-service tend to share a habit: they write the evals before they write the prompts. Everything else follows from that.
None of this guarantees a clean story. OpenAI could ship a model next month that rearranges the assumptions in this piece. But the direction of travel, for now, is clear enough to plan around.