Finance & AI Automation Intern (Remote Part-time)

Data quality is the bottleneck of the AI era. Every Fortune 500 is trying to solve it. We're the team they call. Soda is the data quality layer for Disney, Ralph Lauren, CBRE, HelloFresh, 2K Games, and Nubank. The category is growing rapidly because AI doesn't work on bad data. Behind the product sits a group of three companies in two currencies, and a small finance team that keeps all of it correct. We're hiring a Finance & AI Automation Intern to run core finance processes and automate the manual work out of them. About Us Soda's mission is to monitor the world's decisions. We're building the first Data & AI performance monitoring platform that spans data collection to automated decision-making. Soda helps customers find, understand, and fix data quality issues by combining automated post-production monitoring with pre-production pipeline data testing. Soda was founded in 2019 by Maarten (employee #5 at Collibra) and Tom (creator of multiple open-source projects generating hundreds of millions in value). We commercialize through Soda Cloud and we're well-funded, with investors like Point 9 Capital, Hummingbird, Singular, and Eurazeo. The Role You'll join the Finance team that covers Soda Data NV (Belgium), Soda Data Inc. (US) and NannyML NV, the Belgian company we acquired in 2025. The work is real from week one: invoice processing, month-end reporting, payroll data preparation, expense management, and intercompany recharges across three entities, two currencies and three accounting systems. The second half of the job is making that work disappear. Finance at Soda builds agentically: Claude and other AI tools are part of the daily workflow, not an experiment on the side. You'll learn a process, run it until you know it cold, and then automate it. Sometimes that means an AI agent on GitHub, sometimes a well-designed formula - you pick the right tool and defend the choice. Accuracy is the bar for both: in finance, 99% right is wrong, so every automation ships with the checks and reconciliations that prove its output. What You'll Do Run core finance processes: invoice processing, month-end reporting, payroll data preparation, expense management, and intercompany recharges. Learn them, run them, own them. Automate them: with Claude, ChatGPT, GitHub and the Soda MCP, or with smart spreadsheet design when that is the better tool. Build in the checks: reconciliations and controls in everything you automate, so the output is verified before anyone else has to look at it. Document as you go: process documentation that a colleague and an AI agent can both follow. Support the wider Finance team: reporting, analysis, and ad-hoc projects across our Belgian and US entities. Hiring Process 1. Application with a few short questions 2. Recorded video interview (3 questions, in your own time) 3. 20-minute interview 4. Take-home assignment (plan for 5-6 hours) 5. 45-minute deep dive on your assignment 6. Final interview We answer every application, and every candidate hears back within 3-5 days of each step. If it's not a hell yes, it's a no, and we'll tell you quickly and kindly. Requirements • A finance background: studying towards or recently completed a degree in Finance, Accounting, Economics, Business, or a related field. • Proof that you build: you've used AI tools (Claude, ChatGPT, or similar) and/or built automations, scripts, or advanced spreadsheet solutions, and you can show us one and explain how you checked its output. • Accuracy as a habit: you're the person who notices the 780 that should have been 800, and you don't sign off on numbers you haven't verified. • Independent with new tools: you can pick up GitHub, MCP-based integrations, and finance systems without a course. • Proactive: you spot the manual work and come with a plan for it. You'd rather build the workflow than do the task fifteen more times. • Fluent English, written and spoken. • European business hours: fully remote from any compatible time zone, or hybrid if yo