AI-native software
for the teams who
build things.
Sivect builds AI-native software that embeds intelligence directly into the platforms engineering and operations teams already rely on. Krina is live on the Atlassian Marketplace — with more products on the way.
Krina — Mentions and Follow-ups for Jira and Confluence
Krina uses AI to classify every @mention across three dimensions — Action, Urgency, and Impact — and lets you mark items for follow-up with a date so nothing slips.
Inside the model
Three axes turn chatter into one defensible decision.
Cached response
Once classified, your inbox loads in under 200ms. New @mentions are classified in under 2 seconds.
Architecture — GDPR Art. 25
Only the comment you wrote is analysed — no emails, no credentials, no browsing data. Display names in @mentions are included for classification accuracy.
Auto data eviction (default)
Action items are automatically cleared after 90 days by default. Each user can configure 30–365 days in Settings.
Why AI-Native
Engineered AI-native from the ground up.
Most enterprise tools bolt AI on as an afterthought. Sivect engineers it as the foundation — every data pipeline, every inference path, every UX decision starts from AI.
Forge-native by design
Our Atlassian products run entirely inside Forge — Atlassian's own serverless infrastructure. No external servers, no new vendor approvals, no additional IT security review. Your data stays where it already lives.
Sub-200ms cached intelligence
Intelligent caching and optimised inference deliver classified inbox items in under 200ms. New @mentions are classified in under 2 seconds. No waiting. No manual refresh.
Privacy by architecture
Our 8-layer PII sanitisation pipeline is structural — it runs before every AI call, not configurable after the fact. Emails, phone numbers, API keys, and customer data are replaced with typed tokens before any text reaches the model. Atlassian display names from @mentions are included for classification accuracy. Raw comment data never leaves your Atlassian instance. This is Privacy by Design Principle 1 — the framework now codified in GDPR Article 25.
Philosophy
Intelligence,
embedded.
Enterprise tools generate more information than humans can process. The result is missed signals, delayed responses, and execution gaps that compound across teams and quarters.
Sivect builds AI inference systems that embed directly into existing workflows — classifying, prioritising, and routing information so teams can act on decisions, not drown in triage.
We're starting with Atlassian. We won't stop there.
Every product we build starts the same way: with the research. Before writing a line of code, we map the problem to published work in human-computer interaction, cognitive science, AI, and privacy engineering. Every design decision has a foundation. We don't ship opinions. We ship evidence.
Ready to reclaim your team's focus?
Krina gives every team member a personal, AI-classified inbox for every @mention across Jira, Confluence, and Jira Service Management. No noise. No missed action items.