2 days ago
Amsterdam, NetherlandsEntry Level / Mid Level
Responsibilities
- Configure and tune DTA2 across document types and fields to improve extraction rates.
- Design, build, evaluate, and deploy LLM-based extraction, including prompt design, model selection, guardrails, evaluation sets, and production rollout.
- Build and maintain benchmarking processes to measure extraction improvements, regressions, and causes.
- Identify and implement engineering improvements across DocStreet, DTA2, and Orbis.
- Collaborate with Product, Engineering, Lending Operations, and mortgage-lending clients.
Requirements
- Degree in Computer Science, Software Engineering, Data Engineering, or a comparable field.
- 1–5 years of hands-on technical engineering experience.
- Experience designing and iterating prompts and building benchmarks for LLM-based systems.
- Strong analytical ability, proactive ownership, and the ability to learn new domains quickly.
- Strong communication skills for explaining technical trade-offs to clients and stakeholders.
- Experience with Java and AWS is preferred.
- Dutch language skills are preferred.
Benefits
- Ownership of a strategic 2026 objective in a small, high-visibility team.
- Opportunity to move a production system used by regulated lenders from rule-based to AI-based extraction.
- Work at the intersection of engineering, AI, and the Dutch mortgage domain.
- Hybrid working from the office in Amsterdam.
- Competitive compensation and benefits.
Categories
About Ohpen
Ohpen builds a cloud-based core banking platform delivered as SaaS for banks and other financial institutions, covering savings, lending, mortgages, investments, and pensions. Its products expose APIs and can be paired with business process outsourcing to modernize back-office operations and launch digital journeys. Founded in 2009 and headquartered in Amsterdam, Ohpen serves customers across Europe.