Artificial Intelligence

Range Let Planners Write the Agents Themselves

Range let financial planners, not engineers, write 24 AI skills with a five-step method that keeps human review before any client sees advice.

StartupHub.ai Staff

By StartupHub.ai Staff, reviewed by Daniel Singer, Editor

2 min read

Matthew Ragone said Range let its financial planners write the agents they use every day, and none of them write code, speaking to AI Engineer.

Range Let Planners Write the Agents Themselves - AI Engineer
Range Let Planners Write the Agents Themselves, AI Engineer

Ragone is a certified financial planner who leads AI adoption for the planning team at Range, an all-in-one wealth management firm.

The team started with AI as a chatbot for polishing client messages, then shifted to agents that draft work while planners are elsewhere. Ragone said the guardrail is absolute: "The agent doesn't send anything. A planner reviews every single recommendation before a client sees it."

Two things had to be true for that to work, Ragone said. The agent had to work where the work lives, not in a separate chat window, and it had to know how Range does the work, not just generic planning.

He laid out five steps his team uses to turn that into a repeatable skill: pick, teach, pack, test, and ship. Pick a frequent, judgeable workflow like cash flow or insurance reviews. Teach it with several real cases and narrate judgment out loud. Pack it as a skill that asks questions before it starts and respects limits like never emailing a client. Test it five to 10 times on real cases, then ship it with an owner and a separate reviewer who keeps it current as processes and tax rules change.

Six months in, Ragone said Range has 24 skills and 17 of them are in planning, and the practice spread to sales, product and marketing without a mandate. That spread mirrors wider adoption patterns, where organizations deploy AI agents across a range of types and tasks and implement them with frameworks such as LangChain and Llama Stack or bring-your-own Python, according to NVIDIA. The bottleneck now is not whether a model can draft an analysis, Ragone argued, but who writes down the firm’s method, who tests it and who owns it when 15 other people depend on it.

Sources

Get the free daily brief every morning

Every morning the day's AI news, new AI startups and new funding rounds. Free and takes a few minutes to read, unsubscribe at any time.

Join 64,000 other subscribers.

TopicsRangeMatthew RagoneAI EngineerClaude

StartupHub.ai StaffDaniel SingerWritten by StartupHub.ai Staff, Editorial teamReviewed by Daniel Singer, EditorAbout

StartupHub.ai Staff, Editorial team

The staff writers of StartupHub.ai, ranging from investment analysts to avid AI tool users, early adopters and critical enthusiasts. Backgrounds span engineering, business and the arts. We hold every piece to rigorous standards of research and review.

Daniel Singer, Editor

Daniel Singer is an AI expert who has been writing about, covering and building AI technology since 2018.

More from Daniel Singer

Test your Agent Readiness now

Paste your website. How well ChatGPT, Claude and the other agents can read and use a site, in about a minute.

Watch

The videos behind the news, one flip at a time

The feed mixes the videos from our articles with startup demos and you flip through them like TV channels for free with no signup.