Ai-Driven Finance Platform Engineer
Role details
Job location
Tech stack
Job description
scale our AI-Driven Finance Domain to the next level. Someone who gets involved from problem definition through to measuring whether the thing shipped actually helped, not someone who picks up a ticket, writes the code, and moves on. We want people who are curious about why a feature exists, not just how to build it.Finance is one of the fastest-growing domains at Factorial. Its mission is to be the financial backbone of the companies we serve, handling everything from expenses and invoices to payments, cards, accounting, and analytics so that teams can stop chasing paperwork and processes and focus on the decisions that actually move their business.They're building an AI platform that does the work for you: instead of processing invoices, you oversee a system that reconciles them, negotiates amendments with vendors, and closes out the payment cycle on its own. We're extending our banking-as-a-service platform so companies don't just pay with cards and transfers but also get paid
Requirements
through products like accounts receivable. And we're turning Finance into an engine for cost reduction: planning headcount, policing spend, catching anomalies in invoices and contracts before money goes out, and surfacing the true cost of projects to protect margins. It's a domain where the work has a direct, measurable impact on every company that uses us and where there's plenty still to build*This is not a maintenance role. We're moving fast, the domain is expanding, and there's real engineering work to do.About the teamThe Factorial engineering team is the engine behind Factorial, crafting the tech that makes HR simple, fast, and reliable for thousands of companies.If you love solving hard problems and want to build something that truly matters, build it with them.The core stack is Ruby on Rails and React because the interesting work right now is what we're building on top of that: agentic workflows, LLM-integrated features, systems that behave probabilistically rather than deterministically.What are we looking for?You've shipped AI-powered features and know that evaluating them in production is the hard part, not building them. You think about reliability and user friction before model complexity.You're comfortable reviewing agentic workflows and have opinions about LLM architecture tradeoffs. Not just theoretical ones.You treat AI features as hypotheses. Build, deploy, measure, adjust. You own that whole loop, not just the build part.You can break down non-deterministic outputs into something the rest of the team can reason about and act on.You're comfortable working with Ruby on Rails & React and understand how to integrate AI models into scalable, cloud-native environments.You help teammates who are transitioning from deterministic to probabilistic thinking, not by lecturing, but by working through it together.4+ years in software engineering, with some track record (or serious curiosity) around shipping AI features.You can explain a complex technical decision in English to someone who doesn't write code clearly, without condescension.What You'll Be DoingYou'll spend more time designing systems than writing syntax. That means breaking down complex tasks, defining architecture, and writing high-level prompts that instruct multi-agent systems to generate the underlying code.You review and validate what the agents produce. Security, scalability, structure - that responsibility sits with you, not the model.You'll work on orchestrating multiple AI agents running in parallel, keeping them in sync and making sure the output is coherent across the whole feature.There's no established playbook for a lot of this. You'll define your own workflows, experiment, and share what works - because the team moves fast enough that information silos are a real risk.How We Work (Barcelona / Madrid or Coruña Office)At Factorial, they believe the best products are built when people come together, in person to collaborate, challenge ideas, and move fast. That's w