Speaker profile · Kuala Lumpur
CEO, Trees OS · Founder and AI Fractional Director, Trees Engineering
Fifteen years inside energy megaprojects. Six years building with generative AI, starting in 2020 when there was no ecosystem to copy. He is not an AI vendor. He is an operator who rebuilt his own company around AI, in public, with real payroll and real clients, and now does it for others.
English and French · Keynote, fireside, panel, closed-door briefing, half-day and full-day workshop · Based in Kuala Lumpur, available across ASEAN, travels to Europe
The debate about whether AI tokens become money is the wrong debate. A token behaves like a commodity: priced, stored, hedged, contracted, with states built around it. And its cost structure is a levelised cost of energy in disguise, with a capital asset, a load factor, an electricity price and a unit cost of output.
Which means the people who already know how to model an LCOE know how to model the cost of intelligence. They have simply not been told. The talk closes on the hard ceiling: the watt, not demand, sets the limit.
THE ROOM LEAVES WITHA cost model they already know how to run, applied to a market they thought was foreign.
01 · Why him, and not the other forty
GPT-3 playground in 2020. AI matching shipped into a talent marketplace the same year. ChatGPT from day one. Most people on an AI programme have three years of hands-on. He has six, and the first three were the hard ones, with no ecosystem, no tooling, and nobody to copy.
Trees Engineering is a manpower agency with staff, consultants, clients and an end-of-month payroll. Every AI decision there was made against a P&L, not a slide. Including the ones that failed.
Fifteen years in energy means he can model a levelised cost. Token economics is the same spreadsheet with different labels. Almost nobody on an AI stage can do both, and almost nobody in energy has noticed the overlap yet.
He builds live, says so, and shows the invoice. In a market saturated with certainty, that is the differentiator.
02 · Biographies, ready to copy
Quentin Cloarec spent fifteen years staffing energy megaprojects and has built with generative AI since 2020. Founder, Trees Engineering. CEO, Trees OS.
Quentin Cloarec is a French entrepreneur based in Kuala Lumpur with fifteen years in the energy industry, staffing and coordinating manpower on megaprojects from tender to operations. He has built with generative AI since the GPT-3 playground in 2020, when he launched an AI-matching talent marketplace. He is Founder of Trees Engineering and CEO of Trees OS, and has delivered AI training for a European energy supermajor and a global drilling equipment manufacturer.
Quentin Cloarec is a French entrepreneur based in Kuala Lumpur. For fifteen years he has worked inside energy megaprojects, supplying and coordinating technical manpower across the full cycle: tender, proposal, mobilisation, operations. He founded Trees Engineering, a manpower agency serving the energy sector, and in 2020 launched a talent marketplace built on AI matching, using the GPT-3 playground years before the technology had a market. He is now CEO of Trees OS and AI Fractional Director of his own company, where he ran a full AI transformation on himself first: process audit, redesign, roadmap, and persistent embedded agents in daily operations. He works with clients on the same path, with one rule: adopt what works now, and keep deploying what you will need next. He also created Jaifferson, a salon format that reads a room of six and returns that reading in writing. He speaks about the economics of intelligence, agent governance, and what AI transformation actually costs an operating company.
03 · Track record
Spoken at AI Innovation Days and with La French Tech Malaysia. Corporate AI training on awareness and practical generative AI delivered for a European energy supermajor and a global drilling equipment manufacturer. Advises AI Directors running transformation programmes.
Strong in unscripted formats: panels, fireside, hostile question and answer. Direct, short sentences, named specifics from the field rather than theory. Asks the room open questions and means them.
Already delivered
Energy industry. Manpower on megaprojects: tender, proposal, mobilisation, operations.
Talent marketplace launched with AI matching, built on the GPT-3 playground.
ChatGPT from day one, November.
Full AI transformation of Trees Engineering. Trees OS founded.
Creates Jaifferson, a salon format that reads the room and returns it in writing. Client engagements: training, audit, roadmap, deployment. Advises AI Directors on transformation programmes.

He created and runs Jaifferson, a salon format built on one question worth an evening and never more than six people around it. The conversation is recorded with everyone's consent, and two documents come back: a reading of the room for the table, and a private synthesis that belongs to each participant and to nobody else. Not a transcript, and not a summary.
It matters here for one reason. A speaker who has designed how a difficult room talks, and who reads that room in writing afterwards, is a different proposition on a panel or in a fireside than one who only delivers.
jaifferson.com ↗04 · Workshop
Half day or full day, in company, English or French. Hands-on and role-based. No theory, no tool tourism. Participants work on their own processes and leave with prompts and workflows tied to the job they do on Monday.
What organisers get