Eight companies. Six industries. Every engagement still running.
Construction, financial services, legal, nonprofit, distribution, architecture, manufacturing, coaching. Different problems, different scales, different budgets. Each one started the same way: a CEO or business owner who knew something needed to change but didn't know where to start. Here's what was broken, what I built, and what it cost.
Eight companies across construction, legal, financial services, distribution, architecture, manufacturing, nonprofits, and coaching. Each started with scattered data and manual processes. Same pattern: diagnose, build, measure. Here is what was broken, what I built, and what changed.
Concreto
9-month engagement across 4 phases. Restructured cost tracking from a 10-tab Excel into a real-time engine covering 7 active projects. Built Capataz, a private AI trained on thousands of contract and spec pages with role-based access for legal, technical, and admin teams. Added automated inflation indexing, certification workflows, schedule deviation alerts, and iPad reporting for field supervisors. CEO went from waiting 4 days for a project status to getting it in seconds.
60 min → 30 sec
document lookup
10-15% margin loss prevented
via real-time visibility
A $14B Wealth Advisory Firm
104 documents of institutional expertise were invisible to AI search engines. The firm had a 71x visibility gap between branded and non-branded queries. What started as a 6-week AI visibility project uncovered six operational gaps and became a 9-month Fractional Head of AI engagement. Delivered Voice DNA, 60 authority pages, proposal automation, knowledge systems, intelligence briefing, and team training. Visibility gap closed to 2.1x.
9 months
Fractional Head of AI
6 operational gaps
uncovered and solved
Grupo Lyown
The founding attorney was answering WhatsApp leads during billable hours. The website scored 8 out of 100 and leads from ad spend were going nowhere. Built Victoria, a WhatsApp AI agent that qualifies incoming leads, books meetings automatically, and syncs everything to a centralized database. Meeting booking rate went from 0% to 42% in the first month live.
0% → 42% meeting booking
first month live
60-day build
from kickoff to production
Joy of Impact
The business owner was spending 6-8 hours per grant blueprint and paying a VA to handle repetitive operational work. Configured a complete AI workspace: Voice DNA extraction, 7 pre-built AI skills, organized research, funder databases, and templates. Grant blueprints dropped to 1-2 hours. VA output increased 3-4x with the same headcount.
6-8 hrs → 2-3 hrs
per grant blueprint
3-4x VA output
with same headcount
Rudolph Architecture
Three divisions (Residential, Office, Revalue) with no unified digital presence. No client portal, no CRM, no way for prospects to see past work. Built 3 optimized websites, a client portal with milestone tracking and AI-generated weekly summaries, an admin dashboard connecting CRM, leads, projects, and analytics. Delivered in 4 weeks.
3 divisions unified
one platform, one login
4-week build
2 weeks research + 2 weeks build
Packaging Manufacturer
A packaging manufacturer operating in 4 countries with a website that was barely functional. Field sales carried physical catalogs because the digital tools did not work. Built a multilingual corporate site, 3D product catalog, AI-powered sales tools, a careers portal, and a lead capture system. Sales, catalog, hiring, and inbound leads now run from one system across every market.
4 countries
one system across markets
Field sales team
on a single platform
Almaga
Running a wellness and mystical experiences brand on 7 disconnected tools: Shopify, Systeme.io, WhatsApp, Calendly, and more. Leads fell through cracks between systems. Replaced everything with a unified platform handling memberships, bookings, billing, email, and content delivery under one login. Delivered in 3 months.
7 tools → 1 platform
no more lost handoffs
3-month build
Shopify, Systeme.io, Calendly, and more
Industrial Parts Distributor
Legacy on-premise ERP with tens of thousands of SKUs across multiple US distribution territories. The team spent 65 hours per month building manual reports from Excel pivot tables. Built a real-time sales dashboard pulling directly from the ERP with breakdowns by territory, customer, product, and salesperson. Report generation dropped from 65 hours to 2 hours per month. Delivered in 10 business days.
Hours → seconds
report generation
10 days
kickoff to live dashboard
The pattern
Every one of these engagements started the same way: a CEO, COO, or business owner realized their operation had outgrown its current system. Data was scattered. Processes were manual. Growth was hitting a ceiling because everything required human intervention.
But they all solved differently. Concreto needed real-time cost visibility. Lyown needed lead qualification automation. the firm needed knowledge consolidation. Joy of Impact needed process efficiency. An industrial parts distributor needed their ERP data accessible without Excel gymnastics. But the underlying principle is identical: fix the data layer first, then automate, then govern, then measure.
That's the Custom AI Operating System. Every company I work with gets a different version, built inside their operation, owned by their team. The parts that matter most get built first, in the order that generates the fastest ROI.
The second pattern: speed to working system. Lyown's AI agent went live in 2 months, lawyers stopped checking WhatsApp in the first week, and Victoria reached full autonomy after a 3-month supervised rollout. Joy of Impact cut grant blueprints from 6-8 hours to 2-3 hours in the first week. Concreto took 9 months because the rollout was deeper, but the first dashboard shipped in 6 weeks. I don't believe in pilots. I believe in shipping production systems that create measurable change within 6-8 weeks.
The third pattern: the conversation that starts everything. None of these companies came to me with a fully specified scope. They came with a problem. I diagnosed it, built the roadmap, and executed. That's the AI Ops Audit, a fixed-fee 2-3 week engagement, and you know exactly what's possible.
Frequently Asked Questions
Most are named. Concreto, a $14B wealth advisory firm, Lyown, Joy of Impact, Rudolph Architecture, Almaga. The packaging manufacturer and industrial parts distributor are anonymized at the clients' request. Each case study references production systems that are still running.
Ranges from 2-3 weeks (Joy of Impact) to 9 months (Concreto). The industrial parts distributor dashboard shipped in 10 business days. Lyown's WhatsApp AI agent went live in 2 months and reached fully autonomous operation after a 3-month supervised rollout, for about 5 months total. Most builds produce a working system within 4-8 weeks, the full rollout takes longer if the scope is deeper.
None of these companies came to me with a fully specified scope. They came with a problem. I diagnosed it, built the roadmap, and executed. The AI Ops Audit, a fixed-fee 2-3 week engagement, is how every engagement begins. You get clarity on what's broken and exactly what to build first.
No. Concreto is in Colombia. A $14B wealth advisory firm is headquartered in Miami with a global client base. A packaging manufacturer operates in 4 countries. Lyown is in Colombia. I work remotely with distributed teams and travel when the engagement requires it.
The eight case studies cover construction, financial services, legal, nonprofit consulting, architecture, manufacturing, coaching, and distribution. The common thread isn't industry, it's company size (20-200 employees), manual processes, and data that's scattered across disconnected tools.
See yourself in one of these?
Most of my clients began exactly where you are. Start with a 30-minute conversation or the free AI Ops Audit. I'll ask questions, listen for what's underneath, and tell you what I think makes sense.