Operator-grade software for the work nobody owns.
Kyntra Labs builds operator-grade software for the work nobody owns. Two products in the open right now — Daedalus, for the teams and agencies running Salesforce Marketing Cloud, and Agora, for the cross-functional chaos of running campaigns.
Two products. One opinion about how the work should feel.
Daedalus
Live monitoring and AI recovery for Salesforce Marketing Cloud.
Real-time health across every business unit. An AI agent watches what's running, auto-recovers what breaks, and escalates what it can't — so most routine failures fix themselves before anyone notices.
- Live automation activity across every business unit
- AI recovery — most routine failures resolve automatically
- Incident timeline + history, per business unit
- Built for in-house teams and agencies alike
Agora
A single home for the cross-functional work behind every campaign.
Marketing work lives in a graveyard of Google Docs, Slack threads, Asana boards, and Figma comments. Agora gives that work a home — reusable workflows, live campaigns, and a complete, timestamped change history for every field.
- Modules — reusable forms each team owns and versions
- Blueprints — reusable workflows with if/then branching
- Initiatives — live campaigns with deadlines and ownership at every step
- Field-level change history — every edit timestamped and attributed
I keep building the tools I wish already existed. So I started shipping them.
Kyntra Labs builds the tools that should already exist — the ones teams hack together with spreadsheets, Slack threads, and brittle scripts to solve problems they hit constantly but never have the time or permission to fix properly.
Shared because if they solve it once, they probably solve it everywhere. No ten-year vision deck — just useful software, shipped on a real cadence.
Three layers. No magic.
Plug in. Nothing else changes.
Daedalus installs as a Salesforce Marketing Cloud Installed Package plus a small system resource helper. No agents to deploy, no schemas to migrate, no automations to rewrite. It listens for what's actually running and only surfaces what's relevant.
What's actually running. Not everything ever built.
Live activity for every automation that's run recently — across every business unit. Run history, durations, failure traces. The dashboard the platform never gave you.
Recovery runs before you open Slack.
When an automation fails, Daedalus queues a rerun automatically. If it fails twice, the AI classifies the error, generates a targeted fix, and applies it to the retry — without touching your production automation. You review the diff and promote when you're ready. Or set it to autonomous and it promotes itself.