QX Labs (qxlabs.com) lets you build composable AI agents that use your tools and are synced to your internal data. The platform has three primitives that can work together: Agents (chat-driven, works across Slack/WhatsApp/email, Teams coming soon), Flows (more complex workflows with triggers/guardrails that you can build with chat), and Grids (run your agents over thousands of rows in parallel, like a spreadsheet where every cell can run an agent, script or flow).
Some specifics:
- Built-in knowledge vaults that index and sync internal data (uploads, SharePoint, Granola, working on Google Drive).
- 1,000 tool/app integrations.
- Omnichannel memory - an agent that you're speaking to in WhatsApp can remember and look up a conversation you had earlier in Slack or over email.
- The usual agent harness stuff: scheduling, browser control, persistent workspace, sub-agents etc.
It's free to start with a fixed monthly credit allowance, then you can pay for higher usage. Would appreciate any feedback and happy to answer questions.
hackfather•1h ago
- The biggest adoption blocker was never the tech, it was behavioural: customers didn’t want months of setup and demos/trials mostly because: (1) models + tooling were evolving very fast (decision paralysis) and (2) a lot of enterprise tools required big up-front commitments with no guarantee on uptake/usage. So almost all our large customers started with one use-case, tracked usage, then expanded scope and adoption from there. The question for us became how to create that journey with minimal handholding. We’re now pushing a model of start free -> see results -> scale with usage to expand scope to smaller businesses too.
- Users didn't want a new UI. Usage grew when they could interact with agents from existing surfaces (email, Slack, WhatsApp - whatever fits their style), as it minimised the behavioural friction.
- Grounding in internal data was really important so we spent a lot of time tuning document indexing + retrieval.
- The bespoke, service-heavy model that drove our early growth was sticky but maintenance was a pain given the pace of change (especially as a lean bootstrapped company). Given that many customer were reusing variants of the same underlying primitives + agent harness, we focused on finding ways for customers to update their own agents/workflows and knowledge configurations with minimum friction.
We're still mid-market/enterprise by background and new to self-serve so if something's confusing or clunky let me know.
Jai