Here Comes The Finance Engineer

The Five Stats of the Finance Engineer

Not every finance engineer looks the same. Some are strongest on data infrastructure. Others on shipping automations. Others on audit discipline or systems judgment. 

Below, think of these five stats as the pillars of finance engineering – there’s no single correct allocation, and different builds reflect different strengths.

DATA — If this stat is low, nothing else compensates. You can't build on, control, or measure what you don't understand at the pipeline level. A finance engineer with high DATA knows where a journal entry originates, how it moves through integrations, and exactly where their ERP's data model will fight them.

BUILD — This stat represents building capability, measured in one's ability to ship automations or put matching rules into production. But BUILD without DATA produces brittle things that shatter on edge cases. And BUILD without CONTROLS is an audit risk lying in wait.

CONTROLS — If a machine is going to touch the ledger, it needs discipline baked into the architecture. Developing audit trails, posting scripts into GitHub – these are the elements that a heady finance engineer factors into every build.

STACK — The teams that look smart in two years won't be the ones that built the most. They'll be the ones that built the right things and bought the rest from vendors who've already solved for security, audit, and infrastructure at a depth no accounting team should try to replicate.

ACCT'G-INTEL — You can't automate what you don't understand technically, and you can't evaluate what an agent built if you wouldn't know how to do it yourself. This is the stat that keeps the whole system honest — the reason a finance engineer is still, at their core, an accountant.

The Present Is the Future

The structural barrier that held finance teams back for years is dissolving.

Francisco is clear about why finance teams were late to data infrastructure: "Basically because of the need of engineering support." Every data warehouse, every API connection, every custom report used to require a ticket, a backlog, and someone else's priorities. Now Francisco’s team maintains their own Snowflake instance, their own GitHub repos, their own scheduled data jobs. They built connectors to systems that don't have off-the-shelf integrations — vendor billing APIs, cloud usage data — and they reconcile the results themselves because they're accountants and that's what accountants do. They don't need an engineering team. They are the engineering team.

We're going to see more of this. More Controllers building their own pipelines. More teams owning their data warehouse instead of requesting access to someone else's. More accountants who look at their chart of accounts and their tech stack and their close process and see, for the first time, a system they can redesign — not just a process they execute.

At Numeric, we believe that the accountants who make that transition will be the ones running the profession.

On Becoming a Finance Engineer

If you've read this far and thought that's me — good. If you've read this far and thought "I want to be that" — even better. Here's where to start.

Turn on the terminal.

When we talk with Controllers, they all state the best path forward is starting with a simple problem and trying to automate away.

Here are the most common finance engineering tools for accountants:

• Claude Code
• Claude Cowork
• Codex
• GitHub
• Numeric MCP

You can operate these tools from web, desktop, or via the "terminal" which is the interface for interacting directly with your computer. It's easy to use, and can help you to achieve the fullest capability of MCPs, Claude Code, and the like.

Read what finance engineers are writing.

Operators just like you are producing great content on their builds and experiences:

• Francisco Meyo's The Accountant That Codes — MCP walkthroughs, NetSuite APIs, SQL for accountants, GitHub automation
• Tofique Elder's Substack — field notes from a Controller building with Claude Code
• Alex Altman's Substack — a non-coder documenting his path to a one-click close
• Brock Beyer on LinkedIn — an evangelist for the accounting x coding community

Scan over Numeric's AI Mandate Playbook.

We built it for Controllers and finance leaders navigating the AI mandate with zero margin for error. Data readiness, risk calibration, experimentation frameworks, and what it takes to move automations into the general ledger — grounded in interviews with practitioners doing this work right now.

Start connecting to MCPs from your go-to tools.

MCPs, or Model Context Protocols, allow you to navigate your usual software via AI tooling. Not only can you use individual MCPs to accomplish tasks (ex. using the Gmail MCP to read, write, and send emails via Claude), you can also chain MCPs together for outsized results. Want to build the accruals automation you’ve been needing for months? Link your ERP MCP, the Numeric MCP, and a downloadable Claude Skill to achieve it.

Conclusion

Here comes the finance engineer. Not as a prediction or a conference-circuit talking point — as a person, doing work, today. Building pipelines and agents and dashboards and close workflows that didn't exist six months ago. Finding each other at webinars and in Slack channels and on Substacks. Teaching the person next to them.

The tools will keep changing, and the models will keep improving. But what won't change is the posture: the decision to move from simply being a user of financial systems and to start being an engineer of them.