Data Scientist

Bank OCBC NISP; Jakarta Pusat; On-site; IDR 334,800,000 - 669,600,000; Full time

Job summary

Bank OCBC NISP is seeking a curious ML/AI engineer to own end-to-end AI systems, including LLM workflows, evaluations, and monitoring. You will collaborate with Product, Risk, and Compliance to apply data science to banking scenarios such as credit risk and compliance.

The role emphasizes strong Python fundamentals and experience shipping production ML components, with bonus points for fintech exposure and advanced LLM serving techniques.

Qualifications

• 3+ years of experience building ML or AI systems, including at least one production-ready LLM or RAG system.
• Ability to translate technical trade-offs to non-technical stakeholders (Product, Risk, CX, Compliance).
• Strong Python fundamentals.

Responsibilities

• Own production AI systems and LLM workflows end to end: design, build, deploy, monitor.
• Build the evaluation layer that decides what ships: offline evals, A/B tests, live monitoring.
• Optimize inference with vLLM, prompt optimization frameworks, and fine-tuning (LoRA/QLoRA).
• Work with Product, Risk, and Compliance to apply data science techniques to banking scenarios, including credit risk and regulatory compliance.

Skills

Python

ML/AI systems

Stakeholder communication

Evaluation & testing

Tools

vLLM

Job description

What you'll do:

• Own production AI systems and LLM workflows end to end: design, build, deploy, monitor
• Build the evaluation layer that decides what ships: offline evals, A/B tests, live monitoring
• Optimize inference with vLLM, prompt optimization frameworks, and fine-tuning (LoRA/QLoRA)
• Work alongside experienced teams in Product, Risk, and Compliance to apply data science techniques to real-world banking scenarios, including credit risk and regulatory compliance.

We’re looking for curious and eager-to-learn candidates who are:

• Having 3+ years experiences on building ML or AI systems, including at least one LLM or RAG system you took to production
• Skilled in translating technical trade-offs into actionable insights for non-technical stakeholders (for example: Product, Risk, CX, and Compliance)
• Strong Python fundamentals
• Proven experience in conducting evaluations that directly influenced key product decisions

Bonus points (not required):

• Banking or fintech background
• High-throughput LLM serving with vLLM
• GEPA or similar prompt optimization in practice
• Fine-tuning for financial use cases