Tommy Gao

Tommy Gao

AI Adoption Specialist & Engineer · Christchurch, New Zealand

I turn business problems into reliable systems people use.

15+ years in operations, finance and commercial roles taught me how work actually gets done and where it breaks. Now I design and build systems that fix those problems, and help teams adopt them.

See where I can help How I work
Illustration of colleagues discussing a workflow while another works at a laptop.

What I bring

  • Understand the business

    I uncover the real problem, how people work and the outcome that matters.

  • Choose the right approach

    I assess where AI adds value, where automation is enough and where human judgement must remain.

  • Build reliable systems

    I turn business rules and data into practical tools, with testing and clear controls.

  • Make adoption stick

    I co-design with users and transfer capability through familiar workflows, training and handover.

Where I can help

Manual processes, hard-to-trust data, or AI initiatives that need to become everyday practice: these are the areas where I can contribute.

Finance operations

Reduce repetitive document work and make financial information easier to review and trust.

Explore Finance operations

Commercial pricing

Turn complex pricing rules into consistent decisions, with approvals and traceability.

Explore Commercial pricing

Logistics and supply chain

Translate operating constraints into practical planning tools and clearer decisions.

Explore Logistics

Finance planning and analytics

Make reporting and forecasting repeatable, reconciled and useful.

Explore Finance planning

AI operations and adoption

Move from scattered experiments to workflows people can use, support and govern.

Explore AI operations

Governed data and intelligence

Establish trusted definitions and data foundations for reporting, automation and AI.

Explore Governed data

How I work

  1. 1 Understand

    How people actually work and what the real problem is.

  2. 2 Map

    Process, pain points, decisions, critical steps and quick wins.

  3. 3 Co-design

    Explore options together with the team.

  4. 4 Build

    With a technical team or directly.

  5. 5 Test and refine

    Iterate and learn from exceptions.

  6. 6 Deploy and adopt

    Start with familiar tools and incremental change.

  7. 7 Empower

    Documentation, training, process maps and knowledge transfer.

Selected work

Each project has a full case study: the problem, my contribution, how it works and what changed.

  1. 01 Finance operations AP Intelligence The team reviews validated data instead of re-keying supplier statements, with an audit trail on every edit. In production
  2. 02 Commercial pricing Foodservice Pricing Known pricing errors are designed out by the engine or blocked at an approval gate; the first live round is pending. Built, first live round pending
  3. 03 Logistics and supply chain Freight Optimiser Planners build loads with a rules-driven optimiser in weekly planning, and the business owner signs off every rule constant. In use
  4. 04 Finance planning and analytics Forecast Automation Forecast outputs reconcile to the production reference, and layout changes halt the build before they become wrong numbers.
  5. 05 AI operations and adoption AI Operations Platform Weekly manual triage of the agent workflows replaced by a one-page view rebuilt daily. In daily use
  6. 06 Governed data and intelligence Governed Data Platform A governed Bronze-to-Gold layer, a shared data dictionary and a correction register give reports and AI-generated queries consistent definitions. In place

About

I sit between the people who understand the problem and the people who build the technology. Increasingly, I build both sides.

Background
Roles across operations, pricing, FP&A, commercial analytics and systems, before moving into data and AI.
Current focus
The last two years at the business–technical intersection: data platform, AI workflows, adoption.
With teams
I work alongside the people who do the job, and leave champions behind so the capability outlasts the project.
What I want next
Work where AI adoption and business transformation are the core focus, at a larger scale.
Stack

Data platform

  • Microsoft Fabric
  • OneLake
  • Power BI and DAX
  • PySpark / T-SQL / M
  • SAP BW and ERP sources

Engineering

  • Python
  • pytest
  • Streamlit
  • SQLite / Postgres
  • GitHub Actions
  • Azure Functions

AI tooling

  • Claude Code and skills
  • MCP servers
  • Multi-provider model routing
  • Evaluation harnesses
  • Second-model-family review

Workplace

  • Power Automate and Office Scripts
  • SharePoint and Teams
  • Excel and Outlook automation

Get in touch

If you are hiring for AI adoption or AI engineering work, or planning a system like the ones shown here, I would like to hear from you.

  • AI systems
  • AI adoption
  • Business transformation
  • Finance and operations automation
  • AI operating models
Email Tommy View LinkedIn View SEEK profile Request CV

My CV is available through my SEEK profile or upon request by email.