This is the next instalment in both the StyloAgent and GEPA article series. It continues The StyloAgent Workflow, Part 2 and Evolving Decision Policies with GEPA and Decision Models, bringing persistent agent coordination and evaluated policy evolution…

Behavioural Inference: How I Learned to Stop Worrying and Love Probabilistic Systems

Most software still assumes the world will hand it clean inputs and stable rules. Production usually does neither. This post is the through-line behind a bunch of things I've been building: DiSE, Constrained Fuzziness, CFMoM, Reduced RAG, StyloFlow, The Ten…

Image Summarizer: A Constrained Fuzzy Image RAG Engine

Parts 1-3 described Constrained Fuzziness as an abstract pattern. This article applies those patterns to a working image analysis pipeline that demonstrates the principles in action. CLI Tool - Command-line interface for image analysis Core Library - The…

Constrained Fuzzy Context Dragging: What Gets Remembered

One of the easiest ways to make an AI system fail is to let it remember too much. Most "context handling" approaches boil down to one of two mistakes: Stuff everything into the prompt and hope the model figures out what matters Repeatedly re-summarise…

Constrained Fuzzy MoM: Signal Contracts Over Agent Chatter

Part 1 described a pipeline: Substrate, Proposer, Constrainer. One probabilistic component, bounded by deterministic truth. But what happens when you have multiple proposers? Multiple LLMs? Vision models talking to text models? Heuristics feeding into neural…

Constrained Fuzziness: A Control System Pattern for Probabilistic Components

The Pattern That Kept Emerging It started with my blog's translation tool. The cache kept only 5 tasks per user, with 6-hour absolute and 1-hour sliding expiration. Sixth task? Oldest evicted. After an hour of inactivity, everything expired. Most developers…

Cooking with DiSE (Part 4): Building Systems That Learn, Adapt, and Evolve

Most software architectures assume the system is static. DiSE assumes the system is alive. Note: This is Part 4 in the "Cooking with DiSE" series. See Part 1, Part 2: Graduated Apprenticeships, and Part 3: Untrustworthy Gods for background. This article is the…

Cooking with DiSE (Part 3): Untrustworthy Gods - When Your LLM Might Be Lying to You

In which we discover that your friendly AI assistant might have ulterior motives (and what to do about it) Note: This is Part 3 in the "Cooking with DiSE" series. If you haven't read Parts 1-2, you might want to—though this one stands alone as a slightly…

Cooking with DiSE (Part 2): Graduated Apprenticeships - Training Workflows to Run Without a Safety Net

When your workflows learn to walk without training wheels (and you stop paying for babysitters) Note: This is Part 2 in the "Cooking with DiSE" series, exploring practical patterns for production-scale workflow evolution. Today we're talking about something…