{
    "componentChunkName": "component---src-templates-blog-post-js",
    "path": "/blog/2026-10-11-ai-agents-like-team-members/",
    "result": {"data":{"markdownRemark":{"html":"<p>An AI agent is only useful in engineering if it works like a good team member: it receives a precise task, investigates the file, documents what it found and submits its work for approval. That is how we designed the <strong>autonomous agents</strong> of the TriDyme platform.</p>\n<h2>Agents assigned like employees</h2>\n<p>On the platform, a task is assigned to an agent just as it would be to a colleague: manually, or automatically from a workflow. The agent reads the site, its documents and its equipment, then delivers:</p>\n<ul>\n<li>either a <strong>report</strong>, without writing anything to the portfolio;</li>\n<li>or <strong>proposed values</strong> on a branch, which a human reviews and merges.</li>\n</ul>\n<p>Two separate settings frame its work:</p>\n<ul>\n<li><strong>the profile</strong> sets the AI model used, the (versioned) prompt and the budget per task: cost, number of steps, duration and tasks in parallel;</li>\n<li><strong>the task type</strong> sets what the agent may do: the tools it can call and the actions it can launch. These permissions only change through a reviewed migration, never through a simple setting.</li>\n</ul>\n<h2>Six specialised agents</h2>\n<p><strong>Site documentalist.</strong> Inventories a site’s documents: latest calculation note and latest as-built file (identified by the title-block date), detailed design, equipment sheet, health and safety file, lease; what is authoritative, what is missing, outdated or contradictory, and what to request from the client. It produces a report.</p>\n<p><strong>Site visit preparer.</strong> Writes the preparation sheet for a visit: what the audit does not know or cannot settle (missing data, discrepancies between calculation note and as-built file, foundation, access), the measurements and photos to take, and the safety points. It produces a report.</p>\n<p><strong>Design note reconciler.</strong> Compares the audit’s aerial equipment with the load sheet of the latest calculation note: mounting height, azimuths, operator, status, offset arms, dimensions, weight. It proposes corrections, each one sourced (document and page).</p>\n<p><strong>Rooftop analyst.</strong> Completes and corrects a rooftop audit based on each operator’s reference as-built file: supports, fixings, roof, waterproofing. It never removes a planned item of equipment.</p>\n<p><strong>Structural model reviewer.</strong> When a tower model does not comply, it compares the analysis result with the latest as-built file and calculation note, explains the discrepancy and proposes <strong>the input data</strong> to correct — never a calculated value.</p>\n<p><strong>Audit anomaly fixer.</strong> Resolves rule-check anomalies that deterministic fixes did not settle. What cannot be settled is flagged, not guessed.</p>\n<h2>Safeguards rather than blind trust</h2>\n<p>An agent’s autonomy is only valuable if it is framed. Several principles apply to every agent:</p>\n<ul>\n<li><strong>Nothing reaches the portfolio without human approval.</strong> Proposals go on a branch; an engineer reviews the diff and decides.</li>\n<li><strong>Every value is sourced.</strong> A proposal without a reference document and page is not acceptable.</li>\n<li><strong>AI does not design the structure.</strong> Calculations are still run by the validated deterministic engine (<a href=\"/en/blog/2026-10-11-why-ai-does-not-design-the-structure/\">why</a>).</li>\n<li><strong>Permissions come from the task type.</strong> An agent can only use the tools explicitly allowed for its assignment.</li>\n<li><strong>External connectors are read-only by default.</strong> An agent can use third-party MCP servers only if both its profile and its task type allow it; any write to a third party needs explicit approval.</li>\n<li><strong>Every task has a budget.</strong> Cost, number of steps and duration are capped.</li>\n</ul>\n<h2>Agents at the heart of workflows</h2>\n<p>Agents fit into Forge workflows like any other step: an “Autonomous agent” node assigns a task and waits for its outcome, before moving on, for example, to a human approval and then publication to the client’s system. You can chain document audit, structural analysis, review and correction, with an agent for the cases the rules cannot settle.</p>\n<h2>Key takeaways</h2>\n<ul>\n<li>An agent receives a task, investigates the file and delivers a report or sourced proposals.</li>\n<li>The profile sets the model, prompt and budget; the task type sets the permissions.</li>\n<li>An engineer reviews and merges: the agent proposes, the human decides.</li>\n</ul>\n<p><strong>Give an agent its first task:</strong> discover <a href=\"/en/solutions/agents-ia/\">autonomous AI agents</a> and <a href=\"/en/solutions/workflows-ia/\">AI workflows</a>, or <a href=\"/en/contact/?subject=demo\">talk to an expert</a>.</p>","excerpt":"An AI agent is only useful in engineering if it works like a good team member: it receives a precise task, investigates the file, documents…","frontmatter":{"title":"AI agents that work like team members","language":"en-GB","date":"11 octobre 2026","author":"TriDyme Team","description":"On the TriDyme platform, AI agents are assigned tasks like team members and deliver proposals that an engineer reviews and merges. An overview of the agents and their safeguards.","featuredImage":{"publicURL":"/static/4afbfb9fcc669c9f8c3fc530ae631106/blog-agents-ia-autonomes.jpg"}},"fields":{"slug":"/blog/2026-10-11-ai-agents-like-team-members/"}},"relatedPosts":{"edges":[{"node":{"id":"4f60bf0c-be40-50f2-b64c-138f23fa539a","frontmatter":{"title":"Des agents IA qui travaillent comme des collaborateurs","date":"11 octobre 2026","featuredImage":{"publicURL":"/static/4afbfb9fcc669c9f8c3fc530ae631106/blog-agents-ia-autonomes.jpg"}},"fields":{"slug":"/blog/2026-10-11-agents-ia-collaborateurs/"}}},{"node":{"id":"725ea80a-e400-58d0-adc6-3678ecfe7042","frontmatter":{"title":"Pourquoi l’IA ne calcule pas la structure (et ce qu’elle fait à la place)","date":"11 octobre 2026","featuredImage":{"publicURL":"/static/277a8c96546d448336657f7fe27af91b/blog-ia-calcul-structure.jpg"}},"fields":{"slug":"/blog/2026-10-11-pourquoi-ia-ne-calcule-pas-la-structure/"}}},{"node":{"id":"84b9bead-ae62-58d7-be6e-e2523406c637","frontmatter":{"title":"Why AI does not design the structure (and what it does instead)","date":"11 octobre 2026","featuredImage":{"publicURL":"/static/277a8c96546d448336657f7fe27af91b/blog-ia-calcul-structure.jpg"}},"fields":{"slug":"/blog/2026-10-11-why-ai-does-not-design-the-structure/"}}}]}},"pageContext":{"slug":"/blog/2026-10-11-ai-agents-like-team-members/","locale":"fr"}},
    "staticQueryHashes": ["2349346863"]}