Supernova by Antino
From client conversations to a clear delivery plan.
Supernova brings project documents, client conversations and delivery plans into one workspace. AI agents use those records to draft specifications, plans and updates for your team to review.
For Antino teams and clients
Keep the request, the plan and the decision together.
1. Bring in the source
Start with client documents, calls and project records. Keep the source available when someone needs to check a requirement.
2. Draft the work
Agents turn those records into specifications, screen flows, estimates, tasks and tests.
3. Review and deliver
The team checks the drafts, resolves questions and carries agreed changes into the related work.
Example · repeat an order
“Let buyers order the same items again.”
That request needs a decision: does one tap place an order, or prepare a cart? In our example, one tap adds available items from a previous order to the cart. The buyer reviews changes before placing it.
The result is a requirement, a developer task and test criteria that agree on the same behaviour. This is an illustrative example using fictional project data.
Read the requirement, task and tests →Explore the work from first contact to delivery.
Choose any of the eleven stages. Each shows its input, output and the agents involved. Reviews and updates can repeat during a project; these are connected activities, not a fixed schedule.
The expandable records preserve the homepage’s historical examples, published on 27 August 2026. Sales figures cover multiple accounts; delivery figures follow one anonymised education project. They are not live totals or evidence of time saved. The original collection date for the totals was not recorded.
Stage 1 of 11. Prepare for the first conversation.
Stage 01 · Prospect
Prepare for the first conversation.
Supernova brings together lead records, earlier calls and email threads. Sales agents draft a briefing and suggested follow-up for the account owner to check.
- Input
- Lead and account records, calls and emails.
- Output
- A briefing with relevant history and suggested next actions.
The agents · 6
AlLead researcher
Researches a lead from whatever the records actually hold · and says what they do not hold.
BrainAccount researcher
Researches the account behind the lead from its own history with us.
MultivacDaily briefing
Writes every rep a personal daily briefing: commitments, deals, meetings, one action list.
NestorScrum digest
Merges every scrum call the team held that day into a single narrative.
CalReply writer
Writes the follow-up reply from the thread's own facts.
NorbyQuestion answerer
Answers one-off questions from any page, grounded in the records.
View historical records, dates and reviewers
Published example · 27 August 2026. These records describe activity on the sales accounts or anonymised project, not current totals or measured time saved.
Recorded sales snapshot
22,694
leads
50,945
accounts
133,805
emails
Recorded sales data used for briefings. These counts describe the stored records, not a measured improvement in sales performance.
The person
Rounak Kapoor · Account owner
Opened the account and carried the partnership thread. Present on fourteen of the fifteen recorded calls.
Project dates
2 Jun
First recorded call · a tech-partnership conversation, transcribed in full and kept.
8 Jun
The deal and the account are opened in the CRM, the same morning as a scope call. Source: a client referral.
01 / 11
See what supports a draft.
Source references, calculation inputs and review findings help the team assess the work. They make checking possible; they do not guarantee that an AI output is correct.
- Estimates show the feature matches behind the effort calculation and flag unresolved scope.
- Milestone drafts carry source excerpts. Screen flows link to the use-case steps they cover.
- Powell checks use cases against their source documents. Its findings also need review.
Know who makes the decision.
People decide whether the scope is right, a draft is ready and a change should be applied.
- Reviewers check stage outputs before the work moves forward.
- A feedback round produces proposed updates for confirmation, followed by a check for unresolved notes.
- Repeated corrections can produce a proposed rule. A reviewer decides whether to keep it.
See how Supernova fits your project.
Supernova is used by Antino teams and the clients we work with. Contact us to discuss a project walkthrough or access for your team.
Email your name, company and a brief description of the work you want to discuss. The team can then review the project and the access you need.
Email us about access →Opens your email app. You can also write to hello@antino.ai.
Read access and setup questions