# Engineering | Enrich Labs

> Engineering notes from building Helena, the Enrich Labs AI marketing agent: model fleets, eval harnesses, agent tooling, and production lessons.

_Source: https://www.enrichlabs.ai/engineering_

---

[

![](https://images.unsplash.com/photo-1520869562399-e772f042f422?auto=format&fit=crop&w=1600&q=80)

November 13, 2026

## How we built an omnichannel agent

What we learned bringing Helena to web chat, email, Slack, SMS and iMessage, and MCP hosts: the UX each surface demanded and the plumbing it took.

](/engineering/omnichannel-agent)

[

![](https://images.unsplash.com/photo-1558494949-ef010cbdcc31?auto=format&fit=crop&w=1200&q=80)

August 28, 2026

## When new models ship every week: how we run a fleet of LLMs

How we evaluate, route, and operate multiple LLMs in production: an eval suite mined from real usage, a replay harness, cross-provider adapters, a live-traffic online eval, and a three-gate pipeline for new models.

](/engineering/one-model-to-a-fleet)[

![](https://images.unsplash.com/photo-1518770660439-4636190af475?auto=format&fit=crop&w=1200&q=80)

July 24, 2026

## How we scaled Helena’s tools to tens of thousands

Four architectures we tried for an agent’s tool surface, what broke each one, and where we ended up.

](/engineering/scaling-agent-tools)