BUILT AT THINK & GROW

EngineRoom: AI-assisted event follow-up

EngineRoom is an AI event follow-up app I built at Think & Grow. It scans a badge, enriches the contact from HubSpot, generates a follow-up brief and saves it as a Gmail draft.

This project was built in my role at Think & Grow. This page describes my contribution and only the workflow details I can substantiate publicly.

What it does

EngineRoom is an AI event follow-up app I built at Think & Grow. It scans a badge, enriches the contact from HubSpot, generates a follow-up brief and saves it as a Gmail draft.

The problem it addresses

Event follow-up can become disconnected from the context captured at the event and the history already stored in CRM. EngineRoom addresses the handoff between a newly captured contact, existing CRM context and the first follow-up draft.

My role

I built the event follow-up app and the workflow connecting badge capture, HubSpot context, AI-assisted briefing and Gmail draft creation.

How the workflow works

  1. 01

    A badge is scanned at or after an event interaction.

  2. 02

    The contact is enriched using HubSpot information.

  3. 03

    The workflow generates a follow-up brief from the available context.

  4. 04

    A Gmail draft is created for a human to review before sending.

Tools and integrations

Badge captureHubSpotAI workflowGmail

Example output

The documented output is a Gmail draft plus the context needed for follow-up. A real draft is not published here because no anonymised example has yet been approved for public use.

Results and current status

DOCUMENTED STATUS

Built at Think & Grow. The supplied CV documents the workflow but does not provide measured time saved, send rate, response rate, conversion uplift or revenue impact.

PROPOSED / EXPECTED BENEFIT

The intended benefit is to reduce the friction between an event interaction and a contextual follow-up draft while keeping a human review step before sending. This is the workflow objective, not a measured outcome.

Limitations and lessons learned

A generated draft still needs human review for tone, relevance and factual accuracy. CRM context can be incomplete, and event conversations may contain nuance that is not captured in structured data. Quantified performance data is not yet published.