Risk AI Data Scientist

ING is hiring a Risk AI Data Scientist. Level rates it ; you can apply on ING.

Level's read on this job

On Level's scale, this data scientist role is Builds AI ●●●●: the job is building AI systems.
The posting mentions Mistral, Llama and LangChain among its tools and skills.
Level classes it as a senior role.
ING has 658 other open roles on Level.

AI in this role

Drive the integration of advanced AI capabilities, LLMs, and RAG pipelines into banking risk management.

MistralLlamaLangChainLlamaIndexLangGraphHugging FaceVertex AIPyTorchTensorFlowPythonSQLHugging Face+2

ragai-agentsfine-tuningnlpmachine-learningrisk-managementcredit-risk

ING Hubs Poland is hiring!

The expected salary for this position: 13000 – 22000 PLN

The financial ranges specified in the announcement are adjusted and may differ from the range specified in the remuneration regulations.

We are looking for a Risk AI Data Scientist to drive the integration of advanced AI capabilities into the bank’s overall risk management (out of which Credit risk model maintenance is one of them).

In this role, you operate in the intersection of risk management practices (amongst which credit risk modelling), model lifecycle governance, and advanced AI (LLMs, NLP, Agentic workflows). You will not only build and steer these solutions; you will help design the cognitive layer of the bank’s risk management environment, turning cutting-edge AI into production-ready, compliant solutions.

We are looking for you, if you have:

Master’s degree in mathematics, economics or equivalent,
7+ year experience in risk management (experience with risk modelling is a plus),

Advanced Python, SQL, PyTorch/TensorFlow (SAS is an advantage),

HuggingFace (Transformers, PEFT), LangChain/LlamaIndex, Vector Stores (FAISS/Vertex Search), designing and optimising RAG pipelines,
Experience in manipulating and governing structured and unstructured data for risk management purposes,

Google Cloud Platform (Vertex AI, Workbench),
Strong experience with Azure DevOps (Git, Pipelines),
Experience with end-to-end pipelines (data → model → deployment).

You'll get extra points for:

Experience working in Agile/Scrum teams,

Knowledgeable on AI (risk) governance.

Your responsibilities:

Develop custom models by fine-tuning open-weights models (e.g., Llama, Mistral) on GCP GPUs to understand the specific nuances of risk management in wholesale/retail banking, credit policies, and financial risk (exploration, production and scaling),
Evaluate and monitor GenAI systems in this context (e.g. hallucinations, quality, drifting),
Architect Retrieval-Augmented Generation (RAG) systems to enable interaction with internal policy documents, regulations and other documentation in different formats with high precision,
Work with large-scale unstructured data (documents, PDFs, OCR) as a core part of the role,
Design and implement agentic AI workflows (e.g. LangChain/LangGraph) where AI components plan, reason, and execute multi-step tasks to support risk managers,
Build NLP pipelines to extract complex signals (e.g. transaction patterns, legal clauses) from unstructured text and convert them into usable features,
Write clean, modular Python code in Azure DevOps ensuring models are testable, reproducible, and ready for deployment.

Information about the Team:

The mission of Integrated Risk is focused on providing risk identification, aggregation and insight capabilities at Group level across the various Risk domains. The team department is using those capabilities across the various risk functions, to assume a general oversight of risk governance, policies and frameworks, and to steer group-wide model and implementation activities across locations.

The Bank-wide Credit Risk Models department is responsible for the management of Wholesale Banking (WB) IRB and IFRS9 and the Bank-wide Credit Risk Economic Capital models — including their development, monitoring, and advisory support to the business — in cooperation with relevant stakeholders. All the models in scope are groupwide, managed and developed centrally and consistently applied across all ING’s locations.

The role naming convention in the global ING job architecture will be "Data Scientist IV".

How we rate this

Risk AI Data Scientist at ING rates 90 out of 100 for how much of the daily work is AI. That makes it Builds AI (AI Level 4 of 4). The level is about AI in the job, not seniority.

Classification

Builds AI. The job is building AI systems.

●●●● Builds AI80 to 100
●●●○ Works on AI60 to 79
●●○○ Uses AI40 to 59
●○○○ Little AI0 to 39

Levels come from how often the tools, models and workflows of the role are named in the posting itself. Read the description and count the AI tools, models and workflows it names.

Key facts

Risk AI Data Scientist is an open job at ING.
Level rates this role AI Level 4 of 4: Builds AI.
The role is based in Warszawa (Pańska 97).
The job was posted on July 10, 2026.
Seniority: senior.
Applications are handled by ING. Apply on ING

Prepare for this job

Built from this posting and from similar jobs on Level.

Skills and AI tools this role asks for

RAGAI agentsFine-tuningNLPMachine learningRisk ManagementCredit RiskMistral

Also common in data scientist jobs at this level

Machine learning AWS Spark Deep learning MLOps scikit-learn

Pay

Typical data scientist job: $168k

Use the typical data scientist figure as a reference when pay comes up. See data scientist jobs

Questions you could be asked

How would you design a retrieval step so the model answers from real data instead of guessing?
How do you decide when an AI agent can act on its own versus asking for approval first?
Walk me through fine-tuning a model: what data did you use, and how did you check the result?
What NLP problem have you worked on, and how did you measure whether it actually worked?
Tell me about a project where machine learning was part of your work. What did you do?

Adapt your resume

List these exact terms on your resume: RAG, AI agents, Fine-tuning, NLP, and Machine learning. An applicant tracking system matches the wording, not the idea.
Attach one line of real, concrete experience to at least one of them. A tool named with nothing behind it rarely survives a human read.
Lead with what you built, trained or shipped. This role is judged on the AI system itself, not the tools around it.

Get your CV reviewed before you apply

Send your CV and the role you want next. Your review arrives by email within 3 days.

Get a free CV review Check your resume with the ATS resume checker