Ask in natural language about check-ins, time balance, absences or contracts and get instant answers with charts and tables. Ask it to assign shifts or create schedules: it shows you the change and only applies it when you confirm.
Ask about check-ins, time balance, absences, contracts, schedules and teams just like you would ask a person. The assistant searches your data and answers in seconds, in Spanish or English.
Say "assign Ana the morning shift this week" and the assistant prepares it: creates reusable schedules, assigns different hours per day and even handles night shifts.
Before applying any change, the assistant shows a confirmation card with the exact details of what it is about to do. Nothing is modified until you press Confirm.
Ask for your data the way you want to see it: key indicators, comparison tables, bar charts, trends over time or percentage distributions. Always with your company's real data, never made up.
Turn any chat analysis into a downloadable PDF report with its charts included, ready to share with management, HR or your advisors.
Not sure how something works in JornAda? Ask the assistant: it searches the product documentation and answers with direct links to the exact section of the platform.
It opens from the floating button, on any screen of the panel and in the app too. You ask it the way you would ask someone in admin.
It reads your data with read-only tools and tells you what it is looking at as it goes. You can stop it at any time.
When it helps, the answer comes with a table or a chart, and downloads as a PDF right there. The conversation stays in the history.
Ask it to change a schedule and it does not change it: it proposes it. Nothing runs until you press Confirm.
It looks up employees, clock-ins, the hours balance, absences, sites, teams, work schedules, contracts and assigned schedule, with the same tools the public API uses. Writing, it can do one thing: assign and create schedules and shifts. It does not compensate hours, approve absences, correct clock-ins or add employees. And every change goes through a confirmation card.
Besides your data, it searches the product documentation, so "how do I compensate hours?" has an answer too. It is the first line of Support: a query starts answered by the AI and, if needed, passes to the team with the whole conversation attached. In Support mode it proposes no changes.
The paid plan includes a million tokens a month, and the assistant add-on multiplies that by ten. A question costs around 10,900, measured in production. The header indicator turns red at 90%, and if it runs out the assistant stops answering until renewal: there is no silent degradation.
Company admins only. Employees and team leaders do not see the floating button. The mobile app has the chat, but automated tasks are managed from the web.
No. The only things it can change are schedules and work patterns, plus scheduling automated tasks, always with a confirmation card in front. It does not compensate hours, approve absences, correct clock-ins or add employees.
No. Its scope comes from your session token, not from anything asked in the conversation. And it knows which screen you are asking from, so "why is this day short on hours?" makes sense without repeating who you mean.
It stops answering until it renews, with the calendar month. Before that, the indicator turns red at 90% and a warning goes out once per company per month. Automated tasks do not fail: they are postponed and come back on their own.
No. The turn keeps running in the background and a notification arrives when it finishes; if less than three seconds were left, you are not pinged.
Whatever you would ask someone in admin: who has not clocked in today, who is running up the most hours this month, how a team's hours balance is doing. It can look up employees, clock-ins, absences and holidays, sites, teams, work schedules and contracts, so almost any question about your workforce has an answer. And if it cannot do something, it says so instead of improvising.
Yes. When an answer is worth keeping as a report, it downloads as a PDF from the answer itself, and the text copies with a button. The conversation stays in the side history, with its title set automatically, so you can come back to it days later without having saved it by hand. Each answer can also be rated with a thumbs up or down.
No. The assistant is an admin tool: employees and team leaders do not see the button, and a request made from another role is refused. That is consistent with the rest of the product, where what each person sees depends on their role and their teams. If you want a leader to look up their people's data, they do it from the usual screens.
Yes. It receives which screen you are on, which tab and which record you have in front of you, so "why is this day short on hours?" makes sense without saying whose day you mean. And it keeps the context of the conversation, so you can carry on asking about the same thing without restating the whole question each time.
It tells you, rather than giving you half an answer. And it records the gap for the JornAda team, so what people ask for and does not exist yet reaches whoever decides what gets built. If what you need is help rather than a figure, the same assistant is the first line of Support: the query can pass to the team with the whole conversation.
One question crosses several screens at once. To answer, it can look at clock-ins, the hours balance, absences, contracts and the assigned schedule in the same pass, telling you what it is looking at as it goes. What would be four screens and a separate spreadsheet in the panel is one sentence and one table here. And if the result is worth keeping, it downloads as a PDF from the answer itself.
Ask in natural language about check-ins, time balance, absences or contracts and get instant answers with charts and tables. Ask it to assign shifts or create schedules: it shows you the change and only applies it when you confirm.
