Working notes
from the build floor.
Essays on AI systems, operator tooling, and the gritty edges of shipping software. Written by operators, for operators.
Scoping AI systems the way operators scope everything else
Most AI system scopes fail because they start from the model. We scope them from the workflow, then the model. Here's the template we use.
When internal tools beat SaaS — and when they don't
Every operator hits a ceiling where generic SaaS becomes a constraint. Knowing when to build your own — and when to stay on rails — is the decision that separates operators from tool-shoppers.
Voice agents that actually work: a production playbook
After shipping a dozen voice agents into production, we've converged on a build pattern that handles the edge cases demos never show.
Building marketplaces before you have liquidity
The marketplace literature says chicken-and-egg is the hardest problem. In practice, the hardest problem is shipping v1 without pretending you're past it.
Why we flat-rate everything we productize
Hourly billing makes sense for discovery and aligns nobody for delivery. Here's how we think about fixed pricing, and where it breaks.
AI agents in 2026: what St. Louis operators actually need to know
Forget the chatbot era. Agents that take actions, not just answer questions, are the shift. Here's what's real, what's hype, and where STL businesses should plant flags.
How AI agents actually make money in 2026
Five use cases that pay back inside a quarter, the ones that look good in a demo but leak money in production, and the cost math operators actually need.
AI and the St. Louis construction industry: where the margin actually lives
Procore is not the answer. The GCs winning in STL are using AI for scheduling, procurement, safety, and bid analysis. Here's where the margin hides.
AI consulting in St. Louis: custom systems vs. off-the-shelf tools
Generic AI platforms solve generic problems. STL operators have specific workflows. The difference between buying a subscription and shipping a custom system is where the margin lives.
AI literacy in St. Louis: we can't leave the north side behind
AI adoption in STL is running on two different clocks. Clayton and Ladue are moving. North city is not. The economic cost of that gap compounds every quarter.
Most STL businesses are measuring AI ROI wrong
Time saved is the wrong primary metric. Here's the four-metric framework we use to tell clients whether an AI system is actually paying for itself.
Why custom AI chatbots beat templates for lead generation
Template chatbots crash on the first real conversation. Custom chatbots know your business, score leads intelligently, and route dynamically. Here's the math on why it's worth the build.
St. Louis logistics is sitting on a gold mine of untapped AI
The most strategic logistics hub in America is running dispatch and maintenance decisions on 1990s logic. Here's what predictive routing, maintenance, and HOS compliance actually save.
Missouri government technology: FedRAMP, StateRAMP, and real modernization
Missouri agencies are running mission-critical workloads on infrastructure that's two decades old. Here's what real modernization looks like, phase by phase.
Machine learning for STL local business: what actually works in 2026
Five ML applications that ship cleanly for businesses under 50 employees. Lead scoring, dynamic pricing, document classification, anomaly detection, segmentation that drives action.
Predictive AI for small business: how STL operators make smarter bets
Demand forecasting, churn prevention, cash flow prediction. The three predictive use cases small businesses can actually ship in 2026 without a data team.
The real cost of cheap software in St. Louis
The $15K app that costs $60K. We rebuild two or three of these a quarter. Here's how to evaluate a dev partner so you don't end up in the rebuild pile.
The best tradespeople in STL are invisible online. That's a tech problem.
The discovery layer for trades is broken. Yelp is pay-to-play, Angi is a lead-gen trap, Google is gamed. The best plumbers and electricians in St. Louis are running on word-of-mouth alone.
How to choose a software development partner in St. Louis
Hourly billing is a trap. Full-stack overpromises are another. Here's the difference between a code shop and an engineering partner, and how to tell which you're hiring.
The city-county split is holding St. Louis back from a real tech economy
Two governments, two incentive structures, two workforce pipelines. The split is the single biggest structural drag on regional tech growth. Here's what it's actually costing.
St. Louis is losing its best tech talent and everyone is pretending it's not a problem
Four structural causes nobody wants to own. What the institutional response has actually been. What would actually reverse it. The honest bottom line.
Voice AI receptionist: the 24/7 front desk that never calls in sick
Not a phone tree. A real voice agent that books appointments, answers FAQs, qualifies leads, and routes emergencies. Here's the cost math and where it actually works.
Every missed call is a lost customer: why STL restaurants need voice AI
The phone rings during the Friday rush. Nobody answers. The customer calls someone else. Here's what OpenTable and Resy don't solve, and what voice AI does.
Workflow automation for STL small business: saving 20+ hours a week
Zapier is a graveyard for half-built automations. Here's how real workflow automation ships, which processes to pick first, and when to build vs. buy.
AI agency vs. dev shop. The real difference in 2026.
Everybody puts 'AI' on the website. Buyers who can't tell the difference between an operator-led shop and a dev shop wearing AI makeup pay for it twice. Here is how to sort them.
AI phone agents that actually close, not just answer.
Answering is easy. Closing is a different product. Here is what separates the voice agents that move revenue from the ones that sound impressive on a demo.
How we scope a platform build in 72 hours.
First call Monday. Signed SoW Thursday. Here is the exact process we run for platform engagements, what we need from the operator, and what ships with the proposal.