Data Scientist

illumifin Corporation; Hyderabad, Pune District, Bengaluru; Hybrid; INR 1,800,000 - 2,800,000; Full time

Job summary

illumifin Corporation in Hyderabad, India is seeking a senior data scientist to design, develop and deploy fraud detection models across claims, underwriting and policy applications.

You will lead large-scale data analysis, build real-time scoring engines, apply NLP, graph analytics and anomaly detection to uncover fraud patterns, mentor juniors, and collaborate with investigators, claims, compliance, and executives. Based in Hyderabad, India.

Qualifications

• BTech/Masters or PhD in Data Science, Computer Science, Statistics, Applied Mathematics, or related quantitative discipline.
• 612 years of progressive data science experience, with at least 5 years in fraud detection, risk modelling, or fraud analytics in insurance/financial services.

Responsibilities

• Design, develop, and deploy cutting-edge fraud detection models (supervised, unsupervised, and hybrid approaches) to mitigate risks across claims, underwriting, and policy applications.
• Lead large-scale data analysis initiatives to uncover complex fraud patterns, organized fraud rings, and emerging fraudulent practices.
• Partner with business stakeholders (fraud investigators, claims managers, underwriters, compliance teams) to define strategic fraud detection requirements.
• Build scalable fraud scoring engines and integration pipelines for real-time detection across digital channels.
• Explore and integrate advanced AI/ML methods such as graph analytics, NLP (text mining), and network analysis for sophisticated fraud schemes.
• Mentor and guide junior data scientists/analysts in model development best practices, feature engineering, and performance tuning.
• Ensure all models and analytical solutions comply with regulatory, ethical, and governance frameworks specific to the insurance industry.
• Continuously evaluate fraud performance metrics and recommend proactive enhancements to fraud prevention systems.
• Collaborate with engineering teams to operationalize and scale models in production environments.
• Present technical findings and business impact narratives to senior management and executives.

Skills

Python

R

SQL

Communication skills

Education

BTech/Masters/PhD in Data Science or related

Tools

Spark

Hadoop

Databricks

Job description

Role & responsibilities

• Design, develop, and deploy cutting-edge fraud detection models (supervised, unsupervised, and hybrid approaches) to mitigate risks across claims, underwriting, and policy applications.
• Lead large-scale data analysis initiatives to uncover complex fraud patterns, organized fraud rings, and emerging fraudulent practices.
• Partner with business stakeholders (fraud investigators, claims managers, underwriters, compliance teams) to define strategic fraud detection requirements.
• Build scalable fraud scoring engines and integration pipelines for real-time detection across digital channels.
• Explore and integrate advanced AI/ML methods such as graph analytics, NLP (text mining), and network analysis for sophisticated fraud schemes.
• Mentor and guide junior data scientists/analysts in model development best practices, feature engineering, and performance tuning.
• Ensure all models and analytical solutions comply with regulatory, ethical, and governance frameworks specific to the insurance industry.
• Continuously evaluate fraud performance metrics and recommend proactive enhancements to fraud prevention systems.
• Collaborate with engineering teams to operationalize and scale models in production environments.
• Present technical findings and business impact narratives to senior management and executives.

Preferred candidate profile

• BTech/Masters or PhD in Data Science, Computer Science, Statistics, Applied Mathematics, or related quantitative discipline.
• 612 years of progressive data science experience, with at least 5 years in fraud detection, risk modelling, or fraud analytics in insurance/financial services.
• Strong knowledge of insurance operations and fraud schemes (claims fraud, application fraud, premium leakage).
• Proven expertise with statistical models, ensemble learning, anomaly detection, graph databases/analytics, and NLP.
• Proficiency in Python, R, and SQL, with experience using big data and distributed platforms (Spark, Hadoop, Databricks).
• Experience deploying models into production environments with cloud ML platforms (AWS Sagemaker, Azure ML, GCP Vertex AI).
• Strong communication skills, with ability to influence strategy decisions at both technical and executive levels.
• Track record of delivering measurable business value through fraud loss reduction.