Blog

Title Icon

The Coverage Problem: Why LLMs Can't Do Matrix Math

Submitted by jordan on Wed, 11/04/2026 - 16:00

LLM limitations are not about intelligence — they are about guarantees. A large language model can produce a plausible answer for almost any single input you hand it. What it cannot do is guarantee correctness across the full combinatorial space of inputs and conditions that a real production system must handle. That gap — between "works on the inputs I tested" and "works on every input that will ever arrive" — is the coverage problem.

Title Icon

The Best Agentic AI Tools Aren't the Ones You License — They're the Ones You Actually Run

Submitted by jordan on Wed, 11/04/2026 - 16:00

When someone asks which are the best agentic AI tools on the market, the honest answer is not a ranked list. It's a follow-up question: what problem are you trying to run on them?

Title Icon

Plan in the Cloud, Build Local: The Token Economics of AI Agentic Workflows

Submitted by jordan on Mon, 11/02/2026 - 16:00

The teams running the most cost-effective AI agentic workflows today are not the ones with the biggest compute budgets — they are the ones who are deliberate about which model does which job. Expensive frontier reasoning for planning, cheap deterministic execution for building: that split is the lever that separates a lean agentic operation from one that burns budget on autopilot.

This is not an abstract optimization. It is the difference between a workflow that compounds value over time and one that runs hot, costs a fortune, and gets shut down before it ships anything.

Title Icon

The Seven Shapes of Rework in Agentic Salesforce Work

Submitted by jordan on Sun, 11/01/2026 - 16:00

Rework rarely announces itself. It shows up disguised as productivity — a revert here, a "fix:" follow-up there, a branch quietly closed and re-cut on a new base. Each looks like a small course correction. In aggregate, they are the single largest tax on a delivery stream, and in an agentic one — where an AI author ships fast and there is no default human reviewer standing between the model and the merge button — they compound faster than anyone expects.

Title Icon

The Stale-Baseline Trap: Why AI-Authored Salesforce Delivery Keeps Building on a Fake Org

Submitted by jordan on Thu, 10/29/2026 - 17:00

We recently audited an entire Salesforce delivery stream — 149 merge requests across a single client engagement, start to finish — and asked one question the usual retrospective never asks. Not what shipped, but where a later change had to revert, re-cut, or re-baseline what an earlier one got wrong. About a quarter of the merge requests were rework. And when we grouped that rework by root cause, one pattern sat underneath almost all of it, surfacing in cluster after unrelated cluster.

Title Icon

Three Steps, Then It Compounds: What the First 90 Days of an Agentic Install Looks Like

Submitted by jordan on Tue, 10/27/2026 - 17:00

Most organizations that have studied their AI implementation roadmap arrive at the same realization: the plan looked clean on the whiteboard and stalled within weeks once it met real operations. The tools were fine. The intent was genuine. The gap was the path — nobody had made the first steps concrete enough to execute without heroics.

Title Icon

Done-For-You AI: Why the Managed Agentic Model Beats DIY Agent-Building

Submitted by jordan on Mon, 10/26/2026 - 17:00

Most companies that experiment with AI agents discover the same uncomfortable truth: building them in-house is substantially harder than the demos suggest. The technology works — when context is fresh, tasks are narrow, and someone is watching. The problems surface at scale: agents that hallucinate midway through a workflow, token budgets that spiral, scope that drifts the moment you try to generalize a working prototype into production. Managed AI services solve these problems structurally, not through heroics — and that structural difference is the argument this piece makes.

Title Icon

More Experts, Not Fewer: Why Agentic Scale Demands More Judgment

Submitted by jordan on Thu, 10/22/2026 - 17:00

The standard worry about agentic AI goes like this: agents do the work, experts become redundant, headcount falls. It is a clean story. It is also wrong — and the operators who act on it will find out the hard way.

Title Icon

How to Choose an AI Consulting Partner: A Buyer's Guide

Submitted by jordan on Mon, 10/19/2026 - 17:00

Choosing an AI consulting partner comes down to one test: do they ship a running system you own, or do they sell you slideware and pilots? Evaluate partners on delivery evidence, governance, transparent pricing, and real references. The right partner de-risks the 95% of AI pilots that never reach production — and stays accountable for outcomes, not demos.

Title Icon

The SMB Agentic Buyer Wants Done-For-You AI, Not Enablement

Submitted by jordan on Sun, 10/18/2026 - 17:00

Small businesses buying AI in 2026 want done-for-you AI that runs on their behalf — not training, platforms, or toolkits that require internal expertise they don't have and weren't hired to build.

This distinction matters more than most AI vendors want to admit. The market is splitting cleanly into two buyer types: enterprises that want to build internal AI capability, and small-to-midsize businesses that want the outcomes AI delivers without becoming AI practitioners themselves. Most of the vendor ecosystem is selling to the first group while standing in front of the second.

Pages