small-model orchestration: The Pattern Behind the Best Agents This Quarter
Notes from the teams shipping computer use to real users.
For most of the last year, the conversation around evals-first development has been louder than the evidence. That is starting to change.
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.
Inside Klarna, the rollout looked less like a moonshot and more like a slow migration. A pilot, a champion, a quiet expansion, a budget line.
The skeptical read is that we are watching a feature, not a platform. The optimistic read is that computer use is exactly the kind of feature that becomes a platform when nobody is paying attention.
HubSpot has been quietly running research synthesis through Notion AI for months. The results are unglamorous and, for that reason, more interesting than another benchmark chart.
Inside Linear, the rollout looked less like a moonshot and more like a slow migration. A pilot, a champion, a quiet expansion, a budget line.
The skeptical read is that we are watching a feature, not a platform. The optimistic read is that long context is exactly the kind of feature that becomes a platform when nobody is paying attention.
None of this guarantees a clean story. Mistral 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.