The latest articles on DEV Community by Mattia Costantini (Rancor) (@mattia_costantinirancor). https://dev.to/mattia_costantinirancor https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3801625%2Fb780f2ca-ed25-4412-b5e4-71d186a47e91.jpg https://dev.to/mattia_costantinirancor en Mattia Costantini (Rancor) Mon, 30 Mar 2026 13:35:14 +0000 https://dev.to/mattia_costantinirancor/best-practice-agentic-project-strategy-b1j https://dev.to/mattia_costantinirancor/best-practice-agentic-project-strategy-b1j <p>Hello everyone,</p> <p>Over the past few months, I've been developing an <strong>architectural model</strong> that collects and synthesizes best practices for creating a software project with a Team of Agent.</p> <p>The idea is to <strong>simplify a collaborative work model between humans and AI</strong>.</p> <h2> The Goal </h2> <ul> <li> <strong>trust</strong> [in the tool and the code it produces];</li> <li> <strong>reduction of intellectual debt</strong>;</li> <li> <strong>maximization of flexibility</strong> [understood as the ability to evolve over time (modification, extension)];</li> <li> <strong>reduction of costs</strong> [both in implementation and scalability]</li> </ul> <p>The model is a conceptual and methodological approach, so it can be applied to any development tool.</p> <p>For those using claude-code, I've created a skill that trains the model and helps developers follow these guidelines.</p> <p>This is a long, iterative process of study, testing, and refinement. I know there are still many points to explore, refine (and fix to).</p> <p>Anyone interested can clone the repo.</p> <blockquote> <p>git clone <a href="https://gitlab.com/rancorow/bam" rel="noopener noreferrer">https://gitlab.com/rancorow/bam</a></p> </blockquote> <p><em>The use of my work is absolutely free.</em><br> <em>Any comments/contributions/suggestions are welcome.</em></p> agents ai architecture softwareengineering