Insights
Problems we have solved.
If you recognize any of these, we should talk. If you recognize more than two, we should talk today.
The 11 Things That Will Break Your AI in Production
12 minThere is a pattern in healthcare AI that plays out the same way almost every time. The team builds a proof of concept. It works in the demo. The investor is impressed. The clinical partner nods. Every...
Read more →Your Website Is Invisible to AI Agents
8 minPeople already trust AI enough to listen. When they ask ChatGPT, Gemini, or Claude for a doctor recommendation — and millions already do — the practices that show up get the patient. The ones that don...
Read more →You Are Using AI Wrong
10 minHere is how most companies use AI: you open a chat window, type a prompt, get a response. When you close the window, everything resets. Tomorrow, the AI has no memory of what you discussed. It does no...
Read more →What Happens When You Replace 4 Tools With One System
7 minAdvanced Psychiatry is a 5-provider behavioral health practice in Las Vegas with two locations. Before we got involved, they were paying for an answering service, a patient portal, a standalone schedu...
Read more →Your AI Agent Has a Security Hole You Have Not Tested For
9 minIn 2005, SQL injection was the attack vector nobody took seriously until it was too late. Developers were concatenating user input into database queries, and attackers were extracting entire databases...
Read more →Your AI Forgets Everything Tomorrow
11 minOpen ChatGPT. Have a brilliant conversation. Close the tab. Open it tomorrow. It has no idea who you are, what you discussed, or what decisions were made. Every AI system in widespread use today reset...
Read more →How to Deploy Agentic AI That Survives Production
8 minRiscent is an AI consulting agency with three pillars: agentic deployment, small-language-model fine-tuning and shipping, and AI memory research. This is the first. An "agent" is a model given tools a...
Read more →Fine-Tuning Small Models That Beat Models 30x Their Size
7 minThe frontier-model reflex is expensive and usually unnecessary. For a narrow, well-scoped task, a small model that has been properly fine-tuned can match or beat a model thirty times its size — while ...
Read more →Giving AI Memory That Compounds
8 minMost AI features today are a thin wrapper around someone else's API — and so is your competitor's. What compounds underneath the wrapper is memory. An agent with no memory starts every session from ze...
Read more →A Vision Model That Measures Catches and Catches Cheaters
6 minCompetitive fishing runs on measurement and honesty, and both are hard to enforce from a phone photo submitted in the field. Two questions have to be answered at once: how long is this fish, really, a...
Read more →Phantom Vault: Open-Source Secrets for AI Agents
5 minYou gave an AI assistant access to your machine so it could actually do things — deploy, call APIs, run commands. The moment you did, every secret it can read became something it can leak: into a conv...
Read more →Each insight describes a real problem we encountered and solved in production. The public page shows you we understand the problem. The full playbook — with exact solutions — is available to clients and collaborators.
Book a call →