Blog
Notes from the workshop.
How we build AI coworkers that hold up inside real companies — and what we learn from the teams that run them.
26 August 2026 · Ujjwal Soni
Permission-aware retrieval: the feature every internal AI search needs and most skip
If your knowledge agent can answer a question with a document the asker couldn't open, you have built a data leak with a chat interface. Here is how we designed retrieval so that never happens.
Read →
22 July 2026 · Tisha Garg
Why we built agents around approvals, not autonomy
Fully autonomous agents make great launch videos and bad colleagues. Our Enterprise Agents are designed around a different assumption: a human should be able to stop, approve and understand every consequential action.
Read →
30 June 2026 · Ujjwal Soni
From bespoke automations to a platform: what changed and what didn't
BrewMy{Code} started by building custom automations for companies one workflow at a time. Here is why we turned that work into a product, and what our early customers taught us.
Read →