Director, ball and Agentic Automation

Position: Director, Moneyball and Agentic Automation
Location: Germany

Overview

About the Company:

Armada is an edge computing startup that provides computing infrastructure to remote areas where connectivity and cloud infrastructure is limited, as well as areas where data needs to be processed locally for real-time analytics and AI at the edge. We’re looking to bring on the most brilliant minds to help further our mission of bridging the digital divide with advanced technology infrastructure that can be rapidly deployed anywhere.

Role

Title:

Director, Moneyball and Agentic Automation

Reports To:

Senior Director, Moneyball

Location:

Remote

The Opportunity

Building the “Brain” of the Armada. We are looking for a Director of Moneyball to help build our data and AI infrastructure from zero to one. You won’t just report on the news, you will help make the news.

We use the term “Moneyball” intentionally. In baseball, it meant finding undervalued assets and winning through statistical arbitrage. For us, it means using rigorous data intelligence to optimize our unit economics, forecast with precision, and deploy AI Agents that act as force-multipliers for our operations.

This is a foundational role. You will work on a blank slate (greenfield), help shape the stack, contribute to hiring the initial team, and partner closely with senior leadership to drive strategy.

The Mission

Your mandate is split into two critical engines:

1. The Truth Engine (66% - BI Agents)

You will help build the “Source of Truth” for the entire company. You will replace intuition with evidence, ensuring that every decision, from pricing hardware to optimizing supply chains, is backed by real-time data. You will contribute to owning the Modern Data Stack (MDS) and the financial/operational models that steer the ship.

2. The Automation Engine (34% - AI Agents)

You will play a key role in our internal AI strategy, moving beyond “chatbots” to Agentic Workflows. You will help build and deploy autonomous agents that can “reason, plan, and act” — automating complex workflows in finance, ops, and customer success (e.g., an agent that automatically reconciles invoices or triages technical support tickets).

What You’ll Do

Build the “Moneyball” Foundation (Data & BI)

• Architect from Scratch:
Help select and deploy the V1 Modern Data Stack. Avoid over-engineering, while prioritizing speed-to-insight.

• Define the Metrics that Matter:
Partner with the Senior Director, Moneyball and the Chief Business Officer to define our “North Star” metrics and unit economics (CAC, LTV, Burn Multiples, Gross Margins).

• Research and procure structured and unstructured data from the market to augment the internal data set.

• Develop and implement business ideas and actions to enhance the business.

Deploy the “Digital Workforce” (AI Agents)

• Identify High-Value Workflows:
Audit internal operations to find repetitive, high-cognitive-load tasks that are ripe for agentic automation.

• Build & Orchestrate:
Contribute to the development of internal AI agents using industry-leading frameworks.

• Human-in-the-Loop Design:
Design systems where AI drafts/proposes actions that humans approve, slowly increasing autonomy as trust builds.

Lead & Scale

• Hiring:
Help recruit and mentor a lean, elite team of Data Engineers, Analytics Engineers, and AI Prototypers.

• Culture of Rigor:
Instill a “write it down” culture where data isn’t just a dashboard—it’s the primary language of the business.

Who You Are

• The “Full-Stack” Analyst:
You can write complex SQL transformations in your sleep, but you can also present insights to senior leadership. You are as comfortable with a P&L sheet as you are with a Python script.

• A Builder, Not Just a Manager:
You are Director level and comfortable being an Individual Contributor (IC) for the first 6–12 months. You don’t need a large team to ship value.

• AI Pragmatist:
You’ve moved past the “hype” of LLMs and are focused on utility. You have experimented with building agents and understand the challenges of context windows, hallucinations, and steering.

• Business First:
You understand that a perfect data model is useless if it doesn’t improve the bottom line. You prioritize business velocity over technical purity.

The "Must-Haves"

•…