Event-driven automation & applied AI

Trevy

Event-driven marketing workflows, high-volume messaging, and CRM-grounded LLM content generation.

  • Backend
  • Distributed systems
  • AI inference
Trevy interface

The context

Trevy combines behavioral data and visual campaign workflows to help teams coordinate high-volume marketing operations. Its AI workflows construct context from customer profiles, behavior, and lead status.

My contribution

  • Built a visual, multi-step automation engine using React Flow and Node.js.
  • Designed Redis/BullMQ-based background processing and event-driven backend services.
  • Engineered messaging infrastructure across SendGrid, Gmail, and Outlook integrations.
  • Developed grounded context pipelines for personalized nurturing and follow-up content.

Challenges & solutions

Coordinating high-volume events and long-running campaign workflows.

Used Redis and BullMQ for background jobs, with a visual builder to represent multi-step campaign logic.

Generating content relevant to each CRM lead.

Constructed LLM context from structured customer information, behavioral signals, and lead status.

Pipeline overview

A simplified view of the workflow.

  1. CRM & behavior
  2. Workflow engine
  3. Redis / BullMQ jobs
  4. Messaging & LLM workflows

Scope & outcomes

Data events / month
35M+
Emails / month
15M+
Automation steps / month
30M+

Data events, automation steps, and emails are different workload measures. Internal client data and code are not published.

Building something related?

Tell me about the problem, the constraints, and what success would look like.

Discuss a similar project

Explore more work

Local AI runtime & agent orchestration

Localbots

A Go-based local agent framework and web studio: a manager coordinates specialist bots through one resident Qwen3 model, with no cloud API required after setup.

Backend & synchronization

Smartlistzero

Go-powered CRM synchronization and performance analytics across 70+ agencies managing approximately 5 million contacts.