Long-horizon work · Premium
Projects
Most AI assistants forget. Close the tab and the thread is gone: the plan, the reasoning, what was already tried. A project is the opposite. It holds the plan, the checklist, the goals and the full running history outside of any one conversation, so work that takes weeks does not have to be re-explained every morning.
The problem it solves
Big work does not fit in one conversation.
Some work is a question you ask and an answer you get. Other work is a programme: researching a market properly, rebuilding how your back office runs, migrating years of records, standing up a new line of business. That work takes more than one sitting, and it does not survive being restarted from scratch each time.
When you hand Inscendo IQ something at that scale, it sets up a project to carry it. The project holds the plan and why it looks that way, a checklist of what is done, in progress and blocked, the short-term and long-term goals, and a dated activity log of what it did, decided, and discovered.
You do not usually have to ask. When work is clearly large and multi-session, it sets one up and tells you where to follow along.
Continuity
A new chat tomorrow is not a new start.
A project's memory does not live inside the conversation that created it. Close the chat, come back next week in a completely new one, and the work resumes with the same plan, the same checklist and the same history behind it.
That is what makes long-horizon work possible at all: deep research that unfolds across many sessions, a build that takes a month, an improvement programme with no end date. Work that runs indefinitely is marked Ongoing rather than being forced to show a percentage that would not mean anything.
Staying in control
Questions get pinned, not buried.
When Inscendo IQ hits something it cannot decide alone (a real question, a risk you should know about, something that needs your permission) it pins that item to the top of the project under Needs your attention. It stays there until you dismiss it, instead of scrolling away into chat history. It keeps working on everything else in the meantime.
Going the other way, every project has a place to leave a note. Whatever you write is delivered the next time the project is worked on, even if that is a different session days later. That is how you answer its questions, change direction, or add to the brief between conversations.
It keeps moving without you
Custom agents can carry a project between conversations.
A project does not only advance while you are talking to it. Your custom agents can work a project and update it on their own triggers, so it progresses between your sessions and everything they do lands in the same activity log.
Inscendo IQ usually offers this when it sets a project up, and it comes as one complete proposal: which agents it recommends, what each will do, how often they run, what they will cost (it picks the cheapest capability level that genuinely fits each job), and the risks. You approve the arrangement once.
Agents can work and update a project, but they cannot start new ones. Beginning a project is always a decision made in conversation with you.
Access and lifecycle
Visible to your team, restrictable when it matters.
By default anyone on your team who can use the chat can open the projects panel for your company. If a project is sensitive, the account owner can restrict it, and a restricted project disappears for everyone else, including administrators.
Finished projects are marked complete with a closing summary and the whole history stays readable. Deleting a project is permanent and erases the plan, checklist, goals, notes and history for good.
Creating new projects is part of Premium. If a subscription lapses, every existing project stays fully available to follow, work and finish; only starting brand-new ones waits.
Give it something that takes a month.
Projects are where Inscendo IQ stops being a chat you repeat yourself to and becomes something that carries a programme of work for as long as it takes.