Sr. Applied Data Scientist / ML Lead

• Join the Legacy of Innovation with FCI-CCM:

Backed by a legacy of innovation since 1959, FCI has built over six decades of expertise in communications, continuously evolving to meet changing business and technology needs. By combining deep domain knowledge with innovation, FCI has consistently delivered enterprise-grade communication solutions that help organizations adapt to an increasingly digital world.

Today, FCI continues that legacy through its flagship platform, VARTA, an Insights-Led Customer Communications Platform that empowers organizations to unify customer intelligence, communication workflows, behavioural signals, and engagement orchestration to deliver intelligent, personalized, and real-time customer experiences.

• Broad Function:

FCI is building advanced AI capabilities to help banks convert transactions and behavioral signals into relevant customer engagement and measurable growth.

The initial focus will be on credit cards, with a product designed for reuse across multiple banks and deployment within bank-controlled environments.

The

Senior Applied Data Scientist / ML Lead

will act as FCI’s hands-on analytical lead and will work closely with our banking/data-science co-development partner, along with internal Product, Architecture and Engineering teams.

The role will involve building, challenging and improving machine learning models while creating a lasting analytical capability within FCI beyond the initial engagement.

Roles and Responsibilities (not limited to):

• 1. Applied Data Science & Model Development:

Translate banking growth hypotheses into meaningful features, labels, models and measurable experiments in collaboration with the Product team and banking SMEs

• Build and evaluate approaches for: - Transaction classification

• Behavioural segmentation
• Recommendation and ranking
• Customer propensity and targeting

Select the right approach for each problem, whether simple statistical baselines or specialized machine learning models.

• 2. Model Evaluation & Experimentation:

Own the complete model evaluation framework, including:

• Time-based holdouts
• Leakage prevention
• Class imbalance handling
• Model calibration
• Segment-level error analysis
• Treatment/control evaluation
• Separation of predictive accuracy from incremental business impact

Ensure that model success is measured not through technical accuracy but also through measurable business outcomes.

3. Production ML & Engineering

Write production-quality Python, tests and reproducible training/evaluation pipelines.

Work closely with Data Engineering and MLOps teams on:

• Model deployment
• Monitoring
• Retraining
• Rollback
• Automated validation
• CI/CD for ML workflows

Ensure models are maintainable, reproducible and capable of being operated by other FCI engineers and client teams.

4. Partner Co-Development & Knowledge Transfer

Work hands-on with FCI’s banking/data-science co-development partner.

Reproduce partner results independently, challenge assumptions constructively and document:

• Model assumptions
• Feature definitions
• Evaluation logic
• Rejected alternatives
• Model limitations and risks

Mentor FCI colleagues through practical implementation to build internal capability and reduce dependency on external partners.

• 5. Reusable Multi-Bank Architecture:

Maintain a clean separation between:

• Reusable Core
• Model code
• Feature interfaces
• Evaluation logic
• Training workflowsn and Bank-Specific Components
• Data mappings
• Thresholds
• Configurations
• Client-specific artefacts

Support configurable of the product for subsequent banks without rebuilding the entire solution each time.

• 6. On-Premises / Private Cloud Deployment:

Design ML solutions for bank-controlled environments, including on-premises and private-cloud deployments. Ensure the solution does not rely on continuous public internet access or managed ML platforms.

Build deployment approaches that support:

• Containerized inference
• Offline dependency management
• Secure model execution
• Self-hosted monitoring and feature management

• 7. Model Governance, Security & Privacy:

Work within bank security controls and maintain proper model provenance, documentation and auditability. Support secure handling of sensitive banking and transaction data through:

• PII masking
• Encryption at rest and in transit
• Secrets management
• Access controls
• Secure self-hosted tooling

Support specialist-led assessment of privacy-preserving techniques such as differential privacy and federated learning, wherever required.

• Success in First 6-9 Months: First 90 Days:

• Reproduce partner baseline models.
• Establish evaluation methodology and feature documentation.
• Deliver at least independently implemented analytical improvement.

• By Month 6:

• Operate and improve agreed model workflows.
• Demonstrate reproducibility and segment-level performance.
• Clearly explain model logic and business rationale to Product and Engineering stakeholders.

• By Month 9:

• Lead a controlled model or use-case enhancement.
• Enable at least additional FCI colleague to independently execute critical ML workflows.
• Support reuse and deployment of the solution for a second banking client.

Requirements

• Desired Qualifications & Experience:

• Typically 6–10 years of experience in Applied Data Science / Machine Learning.
• Strong hands-on experience in Python and SQL.
• Practical experience with scikit-learn and modelling frameworks such as XGBoost, LightGBM, CatBoost or PyTorch.
• Strong evidence of personally delivering and maintaining production ML models.
• Experience handling model failures, monitoring, retraining and subsequent improvements.
• Strong understanding of: Statistics, Feature engineering, Experimental design, Selection bias, Confounding, Treatment/control methodologies, Uncertainty and result interpretation
• Experience working with large structured event or transaction datasets.
• Ability to work with noisy labels, missing data and changing customer behaviour.
• Experience collaborating with Data Engineering teams on scalable computation.
• Strong ability to review and test code beyond exploratory notebooks.
• Strong documentation, mentoring and stakeholder communication skills.
• Ability to challenge experts constructively and explain analytical decisions to non-technical stakeholders

Benefits

• The company offers a range of employee benefits including:

• Cashless medical insurance for employees, spouses, and children
• Accidental insurance coverage
• Life insurance coverage
• Retirement benefits including Provident Fund (PF) and Gratuity
• ESI*
• Complementary meal coupons
• Company-paid transportation
• Sodexo benefits for income tax savings
• Paternity & Maternity Leave Benefit
• National Pension Saving
• EL encashment
• Sick Leave

• Typically 6-10 years of experience in Applied Data Science / Machine Learning
• Strong hands-on experience in Python and SQL
• Practical experience with scikit-learn and modelling frameworks such as XGBoost, LightGBM, CatBoost or PyTorch
• Strong evidence of personally delivering and maintaining production ML models
• Experience handling model failures, monitoring, retraining and subsequent improvements
• Strong understanding of statistics, feature engineering, experimental design, selection bias, confounding, treatment/control methodologies, uncertainty and result interpretation
• Experience working with large structured event or transaction datasets
• Ability to work with noisy labels, missing data and changing customer behaviour
• Experience collaborating with Data Engineering teams on scalable computation
• Strong ability to review and test code beyond exploratory notebooks
• Strong documentation, mentoring and stakeholder communication skills
• Ability to challenge experts constructively and explain analytical decisions to non-technical stakeholders