Remember when every new data project required filing a ticket? When business analysts couldn't access the data they needed without an engineer? The rise of analytics engineering changed everything, transforming enterprises from data-informed to data-driven. AI has opened the door for finance's "dbt moment."
In the mid-2010s, the entire business world was pushing to be data driven, but the functional teams tasked with defining and executing strategy were caught in a bind.
They were expected to make decisions using data, but they had limited ability to source or model the data they needed (at least, not in a governed and version-controlled manner). Instead, these functional teams were at the whims of an often-overwhelmed central data team.
They were filing request after request for new data and modifications to the transformation logic for existing datasets, and tooling where they could dive deep into the data on their own terms. But while they waited for the data team to respond, business analysts and leaders were forced to make critical decisions with stale, incomplete data. Or even worse, just by gut feel.
Everything changed when new technologies like dbt and Looker went mainstream: instead of needing centralized data teams to fulfill every data request, business analysts could actually model data and derive actionable business insights on their own, leveraging the same source data and data lake platform used by data engineers.
This shift encouraged technically savvy analysts who went out and leveled up their SQL skills. They assumed an entirely new role: analytics engineer, an increasingly common position that bridges the gap between data engineering and business analytics. Forward-looking organizations realized the potential immediately, with many embedding this resource directly within the functional business teams (e.g. marketing; finance).
The impact was remarkable. Business teams significantly shortened time-to-insight, accelerated iteration cycles, and achieved full visibility into (and control over) how critical metrics were generated.
Business teams could make strategic decisions faster and more confidently — no longer wasting time filing tickets, clarifying requirements, or waiting weeks or months for their ask to be prioritized.
The Finance Parallels
Today, finance teams face a very similar challenge. Finance leaders have long lists of processes that need to be automated and can tell you, specifically, the business outcomes from automating them.
But they lack the tooling and permission to act on those opportunities.
They submit requests to the central IT teams who have the technical capabilities but lack the business context. And IT’s never-ending backlog usually means months of waiting for requests to be prioritized, followed by weeks of alignment meetings to get the technical requirements right.
From there, it’s hoping and praying that external factors or strategic business decisions don’t impact the business requirements before development is complete.
Automation Bottleneck
The recognition moment is clear: finance teams are exactly where data teams were before analytics engineering transformed how they operate.
The dependency problem is identical, and the solution follows the same pattern.
The Emerging Role Division: Finance's "dbt Moment"
AI platforms now enable the same organizational transformation for finance that dbt and analytics engineering enabled for data teams. Finance specialists are now able to easily build, test, iterate, govern, and scale intelligent, agentic workflows.
This means the traditional model of finance requesting automation from IT can evolve into a three-way division of responsibilities:
- Controllers remain masters of "what" needs to happen. They define requirements, ensure compliance, and maintain accuracy standards. This mirrors how marketing or operations leaders identify what metrics matter and how they need to be defined.
- Agentic Process Engineers become masters of "how" work gets done. They design intelligent workflows, orchestrate AI agents, and iterate on automation logic. They understand both business context and system capabilities, bridging the gap between requirements and implementation.
- IT teams focus on being masters of "where" everything runs. They provide the platform infrastructure, security, governance, and system connectivity that enables Agentic Process Engineers to build safely within established guardrails. And they provide enterprise-wide enablement, developing best practices and supporting functional teams as they get started.
Embedded Agentic Process Engineer Model
This represents a capability shift rather than simple job creation. Finance teams transform from automation consumers into automation creators, from users of systems into owners of workflows.
What This Paradigm Shift Unlocks
Organizations that embrace Agentic Process Engineers will realize the same strategic advantages that early adopters of analytics engineering achieved.
Finance becomes a growth accelerator. Teams can spin up new operational processes in weeks rather than waiting months for IT to build new product line reporting systems. When market opportunities arise, finance can adapt immediately instead of being constrained by quarterly planning cycles.
Competitive advantage emerges through operational agility. While competitors wait 18 months for custom IT solutions or struggle to adapt vendor workflows to their business, Agentic Process Engineer teams modify processes immediately when regulations change. Successful practices scale rapidly across business units without lengthy implementation projects.
Institutional knowledge stops walking out the door when experts leave. Agentic Process Engineers codify tribal knowledge into intelligent systems that document both what happens and why certain decisions get made. New hires become productive faster because workflows are self-documenting and self-explaining.
Risk reduction scales dramatically. Intelligent processes provide continuous monitoring and 100% coverage instead of quarterly audits sampling 10% of transactions. Compliance violations get detected in real-time rather than discovered months later during audit cycles.
Finance professionals reclaim their time for actual finance work. Senior staff focus on strategic analysis instead of manual reconciliations. Decision-making accelerates because Agentic Process Engineers can both generate real-time insights and immediately act on them.
The Controller role expands in impact. Controllers can accomplish more with fewer resources while maintaining much tighter feedback loops with their operations.
The Skills and Responsibilities of a Agentic Process Engineer
Agentic Process Engineers require domain expertise combined with systems thinking. Their effectiveness comes from understanding business context deeply while thinking systematically about workflow design.
Core capabilities include end-to-end process thinking, identifying bottlenecks and exception scenarios that manual oversight traditionally handled. They develop system orchestration skills, managing multiple AI agents and understanding how different automation components work together.
Technical literacy also matters, though in a different way than traditional coding roles. Agentic Process Engineers need enough understanding to have productive conversations with IT teams and to grasp what AI platforms can and cannot accomplish reliably. They understand system capabilities and limitations despite not being full-on software engineers.
Change management becomes crucial as teams transition from manual oversight to exception management. Agentic Process Engineers help colleagues adapt from hands-on transaction processing and data manipulation to monitoring intelligent workflows that flag items needing human judgment.
Day-to-day work involves designing intelligent workflows that map business processes to AI automation capabilities. They iterate on business logic through safe testing environments, continuously improving automated processes based on performance data.
How to Identify Potential Agentic Process Engineers in Your Organization
Agentic Process Engineers already exist in many organizations. They just lack proper tools and organizational support.
For example, we talked with a controller at a manufacturing company who saw the complexity of his workload increase exponentially after a merger. He had to work within two ERPs simultaneously, so he took initiative and taught himself SQL and basic Python to automate dozens of data extraction and manipulation tasks that he had been executing manually in excel. He became an unofficial Agentic Process Engineer out of necessity.
This kind of entrepreneurial finance specialist is everywhere. Look for:
- The analysts who spend weekends writing VBA scripts to automate data processing within Excel
- The controllers using ChatGPT to generate Python code to avoid download-upload cycles when systems don’t talk to each other
- The product managers with accounting backgrounds running local automation scripts because they cannot access proper infrastructure
These individuals demonstrate natural aptitude for Agentic Process Engineer work, but they face significant constraints.
Some get "promoted" into centralized automation teams where they lose business context over time and inevitably get spread to have a tangible impact on any given function. Others stay in their functional roles, but have to work around system limitations with inadequate tooling that doesn’t scale or provide adequate audit trails.
The Path Forward
Our perspective is that organizations should encourage and empower their AI-savvy operators to graduate from spare-time experiments to solving low risk business challenges with production-grade solutions.
They’ll need to partner with IT to establish a governed, integrated AI platform that enables safe development and iteration, but we think these future Agentic Process Engineers will have the biggest impact by staying within their function, close to the business challenges they’ll solve in due time.
AI-First Operating Model for Finance
The broader organizational shift can and should happen gradually, following the same organic emergence pattern that created analytics engineering roles. You don't even need to officially establish the new role, you just need to give those excited, AI-savvy analysts, associates, specialists, and manager permission to act.
Over time, establishing best practices that make both finance leadership and IT comfortable with business teams owning application logic will be crucial, but that can’t slow you down. Focus on securing tangible short-term wins and cross-pollinate them with adjacent teams to encourage more bottom-up AI adoption.
All it will take is a few scrappy wins to make the case for the Agentic Process Engineer as a formal role. Start by identifying a manual task within a broader process your team owns, and build an agent to automate that. Start building fluency with simple agents before expanding scope to fully automated processes.
The Future Belongs to Operators That Take Control
As was the case 10 years ago, the technology is ready and the early adopters within finance and operations teams are creating compounding value with AI. They might not have the Agentic Process Engineer title yet, but they’re already reclaiming ownership of their teams’ operations and creating real business value.
CoPlane is building the platform for these eager, AI-savvy operators to build intelligent, enterprise-grade solutions to critical business challenges – and we’re helping them deliver results through strategic advisory and hands-on technical enablement along the way.
Ready to start building Agentic Process Engineer capabilities within your team? Email the founders to learn how!