§ Case Files
Anonymized case studies from real agent deployments. Architecture decisions, failure patterns, and lessons learned.
Case
001
◆ Manufacturing · Workforce
A four-agent research system that identifies manufacturing companies likely to need foreign workforce services, from public signals to a ranked weekly list. Prospecting cut from 40 hours to 2 hours per week.
4 agents · 8 weeks · Production since March 2026
View Case StudyCase
002
◆ SaaS · Customer Success
A three-agent fleet that watches new customer behavior, detects onboarding friction as it forms, and hands the CS team personalized recommendations before the customer asks for help.
3 agents · 6 weeks · Staging
View Case StudyCase
003
◆ Social Media · Publishing
Production-grade publishing pipeline. Every state transition writes a database row, signed URLs are minted only at publish time, and nothing goes live without an explicit human approval. Open-sourced.
1 agent · 4 layers · Production · Open source
View Case StudyCase
004
◆ Operations · Email Automation
One shared inbox, up to 100 inquiries a day, one office manager forwarding everything by hand. We deployed an agent that reads every incoming email, routes it to the right department, and attaches a prepared draft reply. Humans stopped sorting mail and started answering it.
1 agent · 6 departments · in production
View Case StudyCase
005
◆ Sales · Outreach
Generic offer emails get deleted on sight. We built an agent that studies every prospect before writing to them, composes an offer around their actual business, and places their own logo on the PDF. The recipient's first thought stopped being "mass mail" and started being "someone did their homework."
1 agent · 40 offers/day · in production
View Case StudyCase
006
◆ Field Ops · Reporting
An inspector walks a site, records a voice note in the parking lot, snaps a handful of photos, and drives off. By the time they are back at the office, the report is written, formatted, and illustrated. Hours of evening typing turned into minutes of review.
1 agent · voice + photo in · in production
View Case StudyCase
007
◆ Sales · US Market
Our client had a product ready for the US and a pipeline of exactly zero. We built an agent that turns an ideal customer profile into a working pipeline: it sources prospects from the Apollo database, researches each one, runs compliant email sequences across US time zones, and prepares LinkedIn touchpoints for the sales team to send. Market entry as a system, not a spray.
1 agent · Apollo + LinkedIn · in production
View Case StudyCase
008
◆ Logistics · Multi-Agent System
Our largest deployment to date. Nine coordinated agents running the operational core of a trucking company: route planning, load sourcing, driver and vehicle management, fuel, maintenance, registrations, and every spreadsheet the company used to run on. The owner asks a question in plain language and gets the answer in seconds, from data that used to live in thirty different Excel files.
9 agents · full logistics ops · in production
View Case StudyCase
009
◆ Real Estate · Monitoring
In real estate, the good listing is gone by lunchtime. We built an agent that watches every major sale and rental portal around the clock, matches new listings against the client's exact criteria, and sends a real-time alert the moment something appears. The client's team went from refreshing browser tabs to being the first phone call.
1 agent · 24/7 watch · in production
View Case StudyCase
010
◆ Services · Local Outreach
A commercial cleaning company had teams in several cities and a simple growth problem: finding the businesses in each city that need them. We built an agent that deep-researches a city's companies, builds a qualified map of who is worth contacting, and runs the outreach by email and SMS. Each new city stopped being a cold start and became a playbook.
1 agent · multi-city · in production
View Case StudyCase
011
◆ Apparel · Design
A clothing manufacturer's entire inventory, every fabric, button, collar, and tie, cataloged as a design vocabulary for an agent. It generates new garment designs composed only from materials the client actually stocks and can actually sew. Not AI art of imaginary clothes. Producible designs, with the bill of materials attached.
1 agent · full asset library · in production
View Case Study