History engine
Keep the history your systems forget.
History enginePlatform
One AI platform on the systems you already run
Platform overview Watch the tourAsk, see and act on one sample companySolutions · Product
Your tracker shows today. nexibi keeps a dated copy of the lists you choose from a system with a web API, so a board can show how things moved: a sprint burndown, open tickets by status, week by week. And ask what sells and who buys it, in your own words.
Each point is a copy kept that night
Your tracker shows today's tickets. A daily copy kept by the history engine lets a board draw how they moved, from the day the copy starts.
Ask the assistant for burndown, burn-up and velocity rows for each sprint, by day, from the same copy, and draw them on a board.
Each new version records what changed, field by field ("status: In review → Done"). A ticket that disappears is marked, not erased.
A ranked list, ready to download.
The biggest buyers of one product line, and the answer says which orders and dates it counted.
Attach the PDF; it names the page.
Counts and totals are calculated from the rows it reads, not estimated by the AI. It reads a whole list, up to 20,000 rows, and tells you when a list is longer. Ask a follow-up, such as "only the Power tools line" or "now by customer", and it carries on from the last answer. Say what you mean by "sold": units or order-line value, ordered, delivered or invoiced. It says which list it used, and you can tell it what you mean.
On Tarnwick's sample tracker, connected through its web API (history and flows read systems with an API, not a connected database).
"How many open tickets are there in Sprint 14, by status?" The answer: 36 open tickets (To do 12, In progress 16, In review 8), with the list.
Say "keep a daily copy of our open tickets and chart them by status". The assistant proposes the copy on a card: the list, every night at 02:00 your time, kept 90 days, counted by status. Nothing starts until you approve it. It measures the list first and checks the first copy before it draws a chart.
Nothing starts until you approve it.
The board shows one column for each copy, and the sprint's burndown line beside it. History starts on the day the copy starts; as the Track history window puts it, "A chart needs one column per capture — twelve weekly columns take twelve weeks."
A company owner builds a flow in the flow designer: every Monday at 09:00 it reads the open tickets from the tracker, sums them up in one paragraph and e-mails the team. Try it first: nothing is sent and nothing is changed.
Product people often finish a board by hand. The designer is built for that.
TRK-412status: In review → Done
TRK-398story points: 3 → 5
TRK-405priority: Normal → High
TRK-377removed marked, not erased
So you know before the demo.
Keep the history your systems forget.
History engineAsk for one or build it yourself.
DashboardsAsk what sells and who buys it, in your own words.
AI AssistantIn the demo we connect one of your systems, and you ask the questions.