Data Scientist (Senior/Staff)

BETTERDATA PTE. LTD.; Singapore; Hybrid; SGD 90,000 - 120,000; Full time

Job summary

A data solutions company in Singapore is seeking a Senior Data Scientist to lead the application and evaluation of synthetic data solutions. The role involves collaborating with clients to design data pipelines and implement privacy-by-design techniques. Ideal candidates will have over 5 years of data science experience and strong knowledge of generative modelling techniques. Flexibility in work arrangements is offered.

Benefits

Flexible time-off arrangements

Flexible work arrangements

Competitive equity packages

Qualifications

• 5+ years in data science or analytics roles.
• At least 2 years embedded in client environments.
• Strong domain familiarity with financial datasets.

Responsibilities

• Embed briefly onsite with client teams to gather requirements.
• Design synthetic data pipelines for client datasets.
• Implement privacy-by-design techniques and ensure compliance.
• Deliver workshops for client teams on synthetic data.
• Drive adoption of synthetic data methodologies.

Skills

Machine learning frameworks (PyTorch, DP-SGD)

Generative modelling (GANs, VAEs, diffusion)

Privacy-enhancing techniques

Domain familiarity with financial datasets

Education

Bachelor’s or Master’s degree in a relevant discipline

Job description

Who Are We Looking For:

We are seeking an experienced Data Scientist (Senior) to join our team. You will spend a short period embedded within our client’s organization but your primary focus will be on applying and evaluating synthetic data solutions against both client-specific use cases and open benchmarks such as Kaggle datasets.

Key Responsibilities:

Client facing Synthetic Data Solutions

• Embed briefly onsite with the client’s data science, risk, or compliance teams to gather requirements, map data sources, and align on use cases.
• Design and operationalize end-to-end synthetic data pipelines for client datasets from ingestion, training, generation to evaluation.
• Implement privacy-by-design techniques (differential privacy, k-anonymity) and ensure compliance with privacy regulations (GDPR, CCPA, PDPA).
• Deliver hands-on workshops and training for client teams on integrating synthetic data into their analytics and ML workflows.
• Manage client engagements, ensuring deliverables, timelines, and satisfaction metrics are met.
• Drive adoption of synthetic data methodologies by showcasing benchmark successes and business impact.

Benchmarking & Model Evaluation

• Apply and benchmark synthetic data generators on public datasets.
• Develop evaluation frameworks and dashboards to measure fidelity, utility, privacy risk and downstream task performance across multiple benchmarks.
• Iterate on model selection and configuration, fine-tuning generators (GANs, VAEs, diffusion) and pipelines to maximize performance on both client data and open challenges.

Qualifications:

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, Software Engineering, Data Science or a related quantitative discipline.
• 5+ years in data science or analytics roles, with at least 2 years embedded in client environments or professional services.
• Experience with machine learning frameworks (PyTorch, DP-SGD) for privacy-preserving ML applications
• Deep knowledge of generative modelling (GANs, VAEs, diffusion) and privacy-enhancing techniques.
• Strong domain familiarity with financial datasets (transactional, risk, regulatory).

Benefits:

• Flexible time-off arrangements
• Flexible work arrangements - work from office at One North or WFH on some days
• Equity eligibility: Competitive equity packages, with grant size evaluated based on the candidate’s experience, skills, and impact.

How to apply:

Does this role sound like a good fit to you?

• We see this first: Submit your application here
• We see this last: If the above does not work, you may email us your CV (pdf format) at [e-mail] the title of the role in your subject
  Indicate your available start - end dates (DDMMYY - DDMMYY)
  Send along links/supporting information that best showcase the relevant things you have built and done