One person. Hands-on. No overhead.#
I am Tao. I run Xcelerent as a solo practice based in Sydney, Australia, working with teams across Australia, Singapore, and the broader Asia-Pacific region. My focus is advisory, valuation, and professional-services teams — the report- and document-heavy businesses where repetitive work piles up.
Founder, Xcelerent | AI-Enabled Outcome Owner
Connect with me on LinkedIn
I spent years inside corporate finance and data roles — the unglamorous centre of how numbers actually get produced. That is where I learned that the biggest wins are rarely “more strategy” and almost always less repetitive computer work done by smart people who should be deciding, not retyping.
If it is one person, what happens to my system?#
A fair question, and the right one to ask before hiring a solo practice. Three answers, and none of them depends on my availability:
- It runs in your environment. What I build is deployable inside your own infrastructure, or a private, isolated instance — and where data genuinely cannot leave the building, local inference means it does not have to. See security and data handling.
- You own what I build. Documentation and handover are part of every engagement’s scope, not an extra invoice. No lock-in: if it cannot be handed over cleanly, it is not finished. That is stage four — Transfer — of how we work, not a clause you have to negotiate for.
- The outputs are yours in editable form. Reports and decks arrive as Office files you finish in your own tools — not locked PDFs, not a portal you have to keep paying for.
“No overhead” is meant to describe my cost base, not your exposure.
What I have shipped before#
Billing automation that survived audit. Commission and revenue models that outlived three reorganisations. Document and data pipelines that connected systems marketing promised were “integrated”. More recently: LLM-backed workflows where the point is not the model — it is the hours returned to the team.
Credentials and tools#
- AI / LLMs — workflow-first engineering: prompts, RAG where appropriate, local inference when the data cannot leave the building.
- MarkLogic — Certified Administrator and Certified Developer.
- Microsoft data platform — certified Azure Data Engineer and Fabric Analytics Engineer.
- KNIME — implementations, migrations from Alteryx and Excel, production scheduling.
- Python and data engineering — glue code, APIs, extraction, testing — whatever gets the job done without gold-plating.
The MarkLogic and Microsoft certifications above are listed with credential details on my LinkedIn profile.
How I work#
- Deploy, do not advise. I am there to put working systems in place — not to rent you my vocabulary.
- Transfer, do not lock in. You should own what I build. Documentation and handover are part of the scope, not an extra invoice.
- Measure, do not assume. If we cannot point to time saved or risk reduced, we have not finished.
Products#
DocMark.md — AI-powered document branding compliance — is a product I build under its own brand (deep navy / teal). It is separate from Xcelerent’s consulting voice, but it comes from the same obsession with documents done properly.
Citaline — a governed AI receptionist for professional practices — is nearly ready. It answers a business line and speaks only sentences a named person has already approved; where no approved answer exists, it says so plainly and hands the caller to a person.
This site stays focused on hands-on workflow engineering.