Results. Not case studies.
Real problems solved for real businesses. Three engagements. Full depth.
Heartland Fleet Cleaning
Two co-founders. One in Pekin, Illinois running field operations. One in Alpharetta, Georgia running sales, administration, and technology. A fleet washing business launching from scratch with zero existing infrastructure — no CRM, no dashboard, no operational systems, no digital presence.
The challenge wasn’t building a website. It was building an entire company’s technology backbone fast enough to support a real launch — with co-founders operating 700 miles apart.
- Operations Dashboard — Custom Node.js dashboard with real-time visibility into jobs, invoices, revenue, and KPIs. Six critical fixes in Phase 1. Hosted on Railway.
- AI Prospect Intelligence — AI operations agent on dedicated VPS handling prospect research and enrichment. 259 qualified prospects segmented by tier, geography, and fleet size.
- CRM & Sales Pipeline — Monday.com-powered CRM with three-tier outreach strategy. Every prospect tracked from first contact through close.
- Business Comms Stack — Dedicated phone, professional email, customer-facing contact form — all operational before the first prospect call.
- Operational Playbook — Seven versions. Chemical supply chain locked. Equipment specced. Partner revenue share documented. Job pricing built.
A two-person startup operating with the technological infrastructure of a company ten times its size. Revenue tracking from day one. Zero onboarding friction. Full remote operations capability.
When Cam and I started Heartland, I was focused on getting the equipment right and locking down our first accounts. I didn’t have time to think about CRM systems or dashboards — I just needed to wash trucks. Korvex built us everything. We went from a business plan on paper to having a full operations dashboard tracking every job, every invoice, every KPI in real time. They loaded 259 prospects into a CRM pipeline before I even had the trailer hitched up. The thing that changed the game was the remote operations setup. Cam runs the entire sales pipeline, the financials, and all the technology from Georgia while I’m running jobs in Pekin. We’re 700 miles apart and it doesn’t matter. They told us they’d save us 30 hours a week in admin work. Honestly, I think it’s more than that. I just show up, wash, log the job, and everything else takes care of itself.
Edge Terminal
A discretionary futures trader executing on ES and MES contracts. Every morning started the same way: manually pulling gamma exposure data, scanning options flow, reading market structure. Ninety minutes of prep. Every single day.
After the trade? A journal entry that said “long at 5420, stopped out at 5415.” No context. No record of why the trade existed. Intuition doesn’t compound. Data does.
- Morning Bias Engine — Pre-market intelligence synthesizing gamma exposure, real-time options flow, and regime classification into a structured directional read. 90 minutes → under 10.
- Real-Time Dashboard — Python Flask app surfacing institutional positioning, key levels, flow anomalies, and regime shifts. A decision-support layer, not another charting platform.
- AI Trading Journal — Every trade linked to market context at entry — GEX regime, dominant flow, morning bias, structural level. Builds a dataset separating skill from luck.
- Regime Classification — Automated identification of market regimes (trending, mean-reverting, volatile, compressed) so the trader adapts approach instead of applying one playbook everywhere.
Discretionary trading transformed into a repeatable, auditable process. Gut feel replaced with structured decision-making that can be reviewed, refined, and improved over time.
I’ve been trading ES and MES futures for a while, and I was decent at it — but I had no system. Every morning I’d spend 90 minutes pulling gamma exposure data, scanning options flow, reading market structure, trying to figure out which direction made sense before the bell. Korvex built me something I didn’t even know I needed. The morning bias engine gives me a structured directional read in under 10 minutes. The AI journal links every trade to the GEX regime, the dominant flow, the morning bias, and the structural level that defined my thesis. After 50 documented trades, I could actually see which setups I was good at, which regimes I was forcing trades in, and where my edge was real versus where I was just getting lucky. The regime classification alone is worth it. I used to trade the same way every day. Now the system flags the environment and I adapt. My worst losses were always in compressed regimes where I was forcing directional trades. I don’t do that anymore.
Independent Futures Trader
Edge Terminal
Commercial Roofing Platform
A commercial roofing contractor running estimating, proposals, and bid management on spreadsheets and manual processes. Every bid invitation required manual review, takeoff, calculation, and proposal generation. 3–4 hours per job.
General contractor relationships — the lifeblood of commercial roofing — managed through memory and inbox searching. The business was winning work despite its systems, not because of them.
- Roofing Estimating Platform — Full estimating app importing STACK takeoff data, auto-calculating materials, labor, and pricing across seven roofing system templates. Proposals in minutes, not hours.
- Automated Bid Intake — Scans Gmail for invitations, pulls details from PlanHub and BuildingConnected, creates project records in STACK, syncs to CRM. 3–4 hours → 5 minutes.
- Analytics Dashboard — Real-time pipeline visibility: win rates by GC, revenue by roofing system, proposal status. 9,000 lines of battle-tested analytics code.
- GC Relationship Management — Every GC tracked with bid history, tier classification, and automated follow-up cadence. The system enforces the discipline busy estimators forget.
White-label ready. Built for multi-client deployment — each contractor gets a custom instance with their own parsing rules, material logic, and branding. Currently live, targeting 30-client rollout.
Before Korvex built the platform, every bid invitation that came in — from a planroom, from a GC email, from BuildingConnected — I had to manually review, do the takeoff, run calculations in a spreadsheet, format the proposal, and send it. Three to four hours per bid. The estimating platform changed that completely. I import the takeoff from STACK, select the roofing system, and it auto-calculates everything. Proposals generate in minutes. What used to be a half-day process is now 30 minutes. But the bid intake automation is what really saved me. That 3-to-4-hour morning routine of sorting through emails? It’s 5 minutes now. The analytics dashboard gives me visibility I never had — win rates by GC, by roofing system, by month. And the GC relationship tracking is the sleeper feature. Every general contractor has a tier classification and an automated follow-up cadence. I used to forget to follow up with people for months. Now the system won’t let me.
Flynt Waters
Waters Roofing & Construction