Customer Stories

Systems in production, stories from the field.

How engagements unfold: the problem, the approach, and what shipped.

From unscheduled lead to paid invoice in days — not weeks.

Terrafusion built Hybrid Asset Solutions a full end-to-end data platform: backbone data engineering plus an agent orchestration harness that automates the back office across the entire lifecycle of a field job.

8 Weeks to first production release
1500+ Jobs processed per month, 12-month average
40% Year-on-year job growth

Key Outcomes

Speed

Lead to paid invoice in days

An unscheduled job lead now runs to invoice and payment in days — a cycle that used to take weeks.

People

30 people redeployed

Back-office OPEX is stabilised: the same team now supervises, labels, and handles front-office customer work — growth without added headcount.

Revenue

Larger contracts supported

High-priority opportunities and quotes are orchestrated automatically, for clients including Grainger and Graystar.

The Challenge

Hybrid Asset Solutions runs field operations across electrical and plumbing work — and the business was fragmented. Every job moved through the back office by hand: invoices raised and reconciled manually, engineers scheduled by phone and spreadsheet, job reports unread unless someone had time, quotes assembled late enough to cost work. The same job lived in half a dozen places, and nowhere at once.

Thirty people were absorbed by that manual processing, and the lead-to-invoice cycle stretched to weeks. Underneath it all, there was no shared model of the business — no agreed definition of a job, a site, an engineer, or a status — so every system and every team described the work differently. Growth meant hiring more administrators, and the back office became the ceiling on the business.

The Approach

01
Backbone Forward-deployed from day one: extracted, loaded, and transformed the fragmented systems into one ontology, defined from first principles — a single operational record the business builds on.
02
Harness A full orchestration layer, built with field-service CRM Big Change: an agent registry of deterministic and LLM-based agents, tool registries, and evaluation frameworks including LLM-as-a-judge — every action invoked, observed, and judged.
03
Operate First production release in eight weeks; running for over a year. The back office has been repurposed around the agents — and the platform now supports contracts well beyond the business's original ceiling.

Zero to one: a multi-agent platform for conversational commerce.

Terrafusion built Rezolve AI's multi-agent platform for conversational commerce — voice, image, and text — from search backbone to voice models to agent memory. Two years in, it runs in production for merchants from Harrods to M&S.

12 Weeks to first production release
5+ Products in production on the platform
2+ Years embedded — ongoing

Key Outcomes

Platform

Merchants live across industries

Products in production for merchants from Harrods to M&S — the same backbone, redeployed each time.

Voice

State-of-the-art voice models

Best-in-class half- and full-duplex voice, fine-tuned for commerce on Microsoft infrastructure — with co-authored research papers on voice.

Trajectory

Zero to one, then scale

POCs became signed clients and live deployments; Rezolve is Nasdaq-listed and becoming the market leader in its space.

The Challenge

Rezolve AI set out to let shoppers search, discover, and buy through natural conversation — voice, image, and text. That takes semantic search across real inventory, agents that remember context across sessions, and voice that feels human.

Off-the-shelf tools fell short: agents forgot context, voice was rigid, and every merchant deployment risked becoming bespoke. To scale, the platform had to be one backbone — configured per merchant, deployed across industries.

The Approach

01
Backbone Agent and tool registries on LangChain and LangGraph, a semantic search backend across merchant inventory, and backbone APIs wrapping it all — one platform, many merchants.
02
Voice & Memory The Terrafusion-owned core: tiered working and episodic memory for multi-turn, cross-session context, cascaded and full-duplex voice, and LLMs fine-tuned for commerce — discovery, drill-down, cross-recommendation, close — configured by sales persona.
03
Evaluate & Ship Evaluation and agent-testing tools built from scratch, with plan–replan–reflect loops. First production release in twelve weeks; POCs became signed, live merchants.

Building something? We should talk.

Start a conversation