Senior Data Scientist (AI Evaluation & Quality)

Finom; Barcelona; On-site; EUR 70,000 - 110,000; Full time

Job summary

Finom is seeking a founding IC for its AI Team focused on quality, evaluation, and telemetry of AI agents powering Finom's internal operations and tools across ~20 core processes.

You will extend evaluation suites, enforce pre-launch quality gates in CI/CD, and work with domain experts to label cases and resolve disagreements. You will build datasets from real user and operational traffic and refine metrics to guide operational decisions.

Qualifications

• 5+ years in Data Science / Applied AI with product-level metric ownership.
• Production LLM experience (RAG, multi-step tool use) in last 1–2 years.
• Fluent Python & SQL: build data pipelines, evaluation harnesses and dbt models.

Responsibilities

• Own and extend offline/online evaluation suites for ~20 internal AI agent processes-datasets.
• Establish pre-launch quality gates in CI/CD before production changes hit production
• Label cases with domain experts and resolve annotator disagreements
• Build test datasets from real user and operational traffic
• Harden statistical methodology: drift, bias, non-determinism, true metric shifts
• Translate quality metrics into operational decisions; weekly syncs with process owners

Skills

Statistical analysis

Hypothesis testing

Confidence intervals

Python

SQL

LLM evaluation

Data storytelling

Education

Data Science experience

Tools

dbt

Claude Code

Cursor

Codex

Job description

• You'll join Finom's AI Team as the founding IC dedicated to the quality, evaluation and telemetry of AI agents powering Finom's internal operations and tools (including Ops workflows and internal AI analytics engines across ~20 core processes)
• Own and extend our offline/online evaluation suites across ~20 internal AI agent processes-datasets (capability + regression), LLM-as-a-judge rubrics, and deterministic checks
• Establish pre-launch quality gates: enforce pass/fail thresholds in CI/CD pipelines before agent prompt, context, or tool changes hit production
• Work directly with domain experts to label cases and resolve annotator disagreement-breaking definition criteria rather than averaging disagreement away
• Build test datasets derived from real user & operational traffic (tickets, internal chats, colleague queries) rather than synthetic edge cases
• Harden statistical methodology: handle judge drift, verbosity bias, non-determinism, and measure true metric shifts vs. noise
• Translate quality numbers into operational decisions: run weekly syncs with process owners to define clear quality vs. cost/latency trade-offs

Statistical Rigor: Applied knowledge of sampling, hypothesis testing, variance analysis, and confidence intervals on noisy metricsProduction LLM Experience: In the last 1-2 years, you have built, shipped, or evaluated LLM-based systems (RAG, multi-step tool use, agents) as a core, primary job responsibility5+ years in Data Science / Product Analytics / Applied AI roles, with sustained product-level metric ownershipFluent Python & SQL: Ability to write clean data pipelines, evaluation harnesses, and dbt transformation models directlyAutonomous Quality Ownership: Proven track record of owning evaluation methodology or analytics for an entire product or end-to-end process (what to build vs. what NOT to build)AI-assisted coding (Claude Code, Cursor, or Codex) is your default daily authoring environment for Python, SQL, and evaluation scripts-not something you occasionally experiment withYou can walk us through concrete work tasks from the last month where AI coding tools accelerated your engineering and data analysis