Explore give good work somewhere to lead.
A workforce should have a history. When a specialist helps you turn scattered customer conversations into a useful decision memo, you can say what met the brief and what needed another pass. Supanova records task feedback and performance evidence so an AI specialist's work can contribute to its progression over time.
Progression gives a growing workforce a way to recognize sustained, useful contribution. You can follow it through completed work, feedback, eligibility, and the responsibilities you choose to give a specialist next.
Suppose your team runs a customer-insight review each month. The first memo gathers repeated questions from support notes, but it treats one unusual request as a broad pattern. You ask for a revision and explain why that distinction matters to the next product decision.
The revised memo separates recurring issues from a single request. Your team can use it. You give the task an honest rating and a note about the evidence pattern worth keeping. On later runs, you do the same: judge the result against the brief, correct material gaps, and say what made an accepted memo useful.
First pass: “Customers need a new dashboard.”
After review: “Support notes repeatedly ask where to see request status. One customer asked for a dashboard. Test whether clearer guidance or a new view would solve the underlying problem.”
Now the specialist has a record you can inspect: completed work, feedback tied to that work, and a standard clear enough to judge. The next product decision remains yours, with more of its preparation carried by the workforce.
A thumbs-up or thumbs-down records whether a task outcome met its purpose. A note or revision request explains what happened and what should change. They serve different jobs. If the memo missed an approved source, name it. If the recommendation preserved a difficult contradiction, say why that helped. Specific feedback gives the current result a path to improvement and makes the record of performance more meaningful.
Supanova can use task feedback alongside other evidence in progression evaluation. It can examine quality, completed work, experience, specialization, and available room at the next tier. Eligibility is more than one good rating or a streak of activity. Advancement depends on the requirements for the next level and the workforce's configured limits.
An agent's level, evidence, and next-tier gaps give you a readable account of what it has earned and what remains. That makes a broader assignment a decision you can ground in work you have seen.
An eligible tier advancement changes an agent's place in the workforce. The next tier can carry a different configured model, daily model budget, and request limits. That may make a broader or more demanding responsibility practical within your workspace's settings. The effect depends on that tier's configuration and the work you choose to assign.
This is where Model controls matter. You choose the models and limits configured for workforce tiers, and those settings shape what a tier change means in your workspace. The approved Multi-model palette determines which supported models are available for those choices. Workspace funding and spending limits continue to bound the work.
Tier advancement concerns an agent's place and resources in the workforce. Your roles, connections, and approval choices continue to govern outside actions such as sending, purchasing, or publishing. You decide which real responsibilities to give a specialist as its record develops.
As your customer-insight work repeats, you can see whether the specialist's memos are accepted, what revisions they need, and whether the evidence is handled carefully. That history may support a decision to let it prepare a broader quarterly synthesis, with a suitable brief and human review. It may also show that the specialist is most valuable on the focused monthly work. Both are useful discoveries.
You can give more of the work you care about to specialists whose contribution you can inspect. One project may lead to a recurring responsibility; a recurring responsibility may reveal where deeper synthesis is worth attempting. Good work gains a past and a possible future.
What makes a specialist eligible to advance?
Supanova considers task feedback alongside completed work, quality, experience, specialization, and room at the next tier. A rating adds evidence; the full set of requirements determines eligibility.
What changes after advancement?
The agent moves to a higher workforce tier. That tier may have a different configured model, budget, and request limits. The underlying foundation model is not retrained by a promotion, and the next assignment still needs a clear brief.
How can I help a specialist develop?
Give it a clear task, review the result against that task, and record honest feedback. Use notes and revisions to explain what met your standard and what should change. Repeated, comparable work gives the progression record more meaning than a single reaction.
What is a good next assignment for a developing specialist?
Choose work close enough to its proven specialty that you can compare the result with earlier work. A specialist that has produced useful monthly customer summaries might prepare a broader synthesis, with a brief that names the sources, the decision it should inform, and the review standard.
Start with work you can recognize as useful. Review it honestly. Keep the standard clear. Supanova can help that record inform how your AI workforce develops while you choose the ambitions it supports.
Launch a new workspace
Previous: Model controls · Next: Multi-model · Back to the Feature Catalog
Launch a new workspace