Inside a Production Agent That Actually Ships With small-model orchestration
A field report on RAG-as-a-service and what it changes for sales teams.
If you spend enough time watching the AI industry, you stop reacting to launches and start tracking patterns. Inside a Production Agent That Actually Ships With small-model orchestration is one of those patterns.
Inside Stripe, the rollout looked less like a moonshot and more like a slow migration. A pilot, a champion, a quiet expansion, a budget line.
Teams that win with fine-tuned distillation tend to share a habit: they write the evals before they write the prompts. Everything else follows from that.
Shopify has been quietly running incident response through Codeium for months. The results are unglamorous and, for that reason, more interesting than another benchmark chart.
Intercom has been quietly running research synthesis through Raycast AI for months. The results are unglamorous and, for that reason, more interesting than another benchmark chart.
The skeptical read is that we are watching a feature, not a platform. The optimistic read is that vision input is exactly the kind of feature that becomes a platform when nobody is paying attention.
Eval harnesses, once an afterthought, are becoming the most important piece of code in many AI projects. Duolingo's team treats theirs the way an SRE team treats a runbook.
None of this guarantees a clean story. Meta FAIR 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.