New opportunity

Lead Data Scientist - IT (Information Technology) - Bangalore, India

Societe Generale · Bangalore, India

About this role

The Lead Data Scientist will spearhead the design, development, and deployment of advanced Machine Learning and Generative AI solutions to address complex business challenges at Societe Generale. This leadership role involves managing the full model lifecycle, from data exploration and feature engineering to production deployment and monitoring. Responsibilities include defining performance metrics, mentoring team members, and driving innovation through AI agents and LLM-based architectures. Key responsibilities: Lead end-to-end AI initiatives, including classification, clustering, anomaly detection, and predictive analytics. Design and implement Generative AI solutions using RAG, embeddings, and multi-agent systems. Establish evaluation frameworks for model reliability, explainability, and AI safety. Mentor Data Scientists and provide technical leadership across the organization. Required qualifications: 9+ years of experience in Data Science, Machine Learning, or Applied AI. Proven track record of leading AI projects from experimentation to production. Strong expertise in statistical modeling, experimental design, and large-scale data analysis. Hands-on experience with LLMs, vector databases, and prompt engineering. Ability to translate business requirements into scalable technical architectures. Preferred qualifications: Experience with AI governance, hallucination mitigation, and integrating AI into enterprise workflows. The role is based in Bangalore, India, and operates on a hybrid work model.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

Machine LearningData ScienceSupervised LearningUnsupervised LearningClassificationClusteringAnomaly DetectionForecastingRecommendation SystemsPredictive AnalyticsStatistical ModelingHypothesis TestingExperimental DesignFeature EngineeringModel ExplainabilityModel MonitoringGenerative AILarge Language Models (LLMs)Retrieval Augmented Generation (RAG)EmbeddingsVector DatabasesPrompt EngineeringAI AgentsMulti-agent SystemsAI GovernanceAI Guardrails

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

End-to-End AI Lifecycle Management
The role requires ownership of the full model lifecycle, from initial data exploration and feature engineering to production deployment and ongoing monitoring. You must demonstrate the ability to define performance metrics and validation frameworks that ensure models remain relevant and impactful in a production environment.
Generative AI and Agentic Architecture
A core priority is designing and implementing advanced GenAI solutions, specifically using RAG, vector databases, and multi-agent systems. You will be expected to integrate these technologies with enterprise workflows, APIs, and knowledge repositories while ensuring scalability and reliability.
Technical Leadership and Mentorship
As a Lead Data Scientist, you are responsible for mentoring team members and providing technical guidance across the organization. This includes establishing best practices for statistical analysis, AI governance, and model explainability to maintain high standards of quality assurance.
AI Governance and Safety
You must prioritize the implementation of AI guardrails, hallucination mitigation, and responsible AI principles. This involves designing evaluation methodologies that measure model accuracy and safety, ensuring all AI initiatives align with enterprise governance requirements.

Likely challenges to think through

Balancing Innovation with Enterprise Constraints
Integrating cutting-edge GenAI and agentic systems into a large, established financial institution often involves navigating strict security, compliance, and legacy system constraints. Prepare by discussing how you have successfully navigated enterprise-grade governance while still driving rapid innovation and pilot adoption.
Ensuring Reliability in Agentic Workflows
Multi-agent systems and LLM-based architectures introduce risks related to non-deterministic behavior and hallucinations. The challenge lies in building robust evaluation frameworks that provide measurable reliability. Prepare by detailing your experience with automated testing, guardrails, and human-in-the-loop validation strategies.
Translating Complex Business Needs into Scalable AI
There is a significant difficulty in bridging the gap between abstract business challenges and technical AI implementation. You will need to demonstrate how you translate high-level operational goals into specific, scalable technical architectures that deliver measurable business value.

Practice your answers

Technical practice

Designing RAG Architectures for Enterprise

What to coverExplain how you would design a Retrieval-Augmented Generation system to minimize hallucinations while integrating with secure enterprise knowledge repositories. Focus on the trade-offs between different embedding strategies, vector database selection, and the retrieval pipeline efficiency.

Multi-Agent System Orchestration

What to coverDescribe your approach to designing multi-agent systems for complex task orchestration. Discuss how you handle agent communication, state management, and error handling when agents interact with external APIs or business workflows.

Model Lifecycle and Production Monitoring

What to coverDetail your strategy for managing the full model lifecycle from experimentation to production. Specifically, address how you implement model monitoring to detect data drift and performance degradation in real-world production environments.

AI Guardrails and Governance

What to coverHow do you implement AI guardrails to ensure safety and compliance in Generative AI applications? Discuss your experience with hallucination mitigation techniques and establishing governance frameworks for model deployment.

Experimental Design and Statistical Rigor

What to coverDescribe a scenario where you had to design a rigorous experiment to validate a predictive model. Explain how you define success criteria, handle hypothesis testing, and ensure statistical significance in your results.

Feature Engineering for Large-Scale Data

What to coverDiscuss your methodology for feature engineering when working with large-scale structured and unstructured datasets. Focus on techniques for maintaining feature consistency between training and inference pipelines.

Model Explainability Techniques

What to coverCompare different model explainability methods (e.g., SHAP, LIME) and explain how you choose the appropriate technique based on the model type and the needs of business stakeholders.

Foundation Model Selection Criteria

What to coverWhat framework do you use to evaluate and select foundation models for specific business use cases? Address the balance between model performance, operational costs, latency, and security requirements.

Behavioural practice

Leading AI Initiatives from Concept to Production

What to coverDescribe a time you led an end-to-end AI project from ideation to production. Focus on how you translated business requirements into a technical architecture and managed the team through the development lifecycle.

Mentoring and Technical Leadership

What to coverProvide an example of how you have mentored junior data scientists or provided technical leadership across a team. Explain how you balanced hands-on technical work with the need to guide others' professional growth.

Navigating Technical Trade-offs with Stakeholders

What to coverTell me about a situation where you had to explain a complex technical trade-off (e.g., model accuracy vs. latency) to non-technical stakeholders. How did you ensure alignment on the final decision?

Driving Innovation in a Regulated Environment

What to coverDescribe a time you identified an opportunity to improve operational efficiency using AI. How did you manage the risks and governance requirements while driving the project to adoption?

Sharpen your knowledge

Topics to prepare

Agentic AI and Multi-Agent Architectures
The role requires designing systems where AI agents perform reasoning, planning, and task orchestration. Review conceptual frameworks for multi-agent systems, including how agents communicate, manage state, and handle hand-offs in enterprise workflows. This is critical for the requirement to build autonomous workflows.
AI Governance and Model Reliability
Focus on the theoretical aspects of AI safety, hallucination mitigation, and model explainability. Study how to establish evaluation frameworks for GenAI that measure accuracy and operational readiness. This aligns with the need to ensure responsible AI principles in a banking environment.

Skills to revise

Retrieval-Augmented Generation (RAG) Implementation
Practice building a RAG pipeline that integrates vector databases with LLMs. Focus on optimizing retrieval accuracy and embedding quality. This is a direct requirement for the hands-on development of Generative AI solutions mentioned in the job description.
End-to-End Model Lifecycle Management
Exercise the full MLOps cycle, specifically focusing on production monitoring, performance evaluation, and feature engineering for large-scale datasets. Prepare to discuss how you have transitioned models from experimentation to industrialization, as this is a core expectation for a Lead Data Scientist.
Prompt Engineering and Guardrails
Develop a set of systematic prompt engineering techniques for complex tasks and design guardrails to prevent model misuse or data leakage. This is essential for the requirement to design and optimize safety mechanisms for enterprise AI applications.

Your preparation checklist

Prepare a portfolio of end-to-end AI project case studies
Document 2-3 projects where you led the full lifecycle from experimentation to production. Focus on your methodology for feature engineering, model monitoring, and how you translated specific business challenges into scalable technical architectures.
Develop a framework for Generative AI and Agentic systems
Create a technical summary explaining your approach to RAG, multi-agent orchestration, and vector database selection. Be prepared to discuss how you implement AI guardrails, mitigate hallucinations, and ensure model explainability in enterprise environments.
Refine your technical leadership and mentorship narrative
Prepare concrete examples of how you have mentored junior data scientists, managed team performance, and established best practices for model governance and quality assurance within a large organization.
Formulate strategic questions for the interviewers
Prepare questions regarding the current maturity of their AI governance frameworks, the primary business domains they are prioritizing for GenAI integration, and how the team balances innovation with operational reliability.

Availability has not yet been verified. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…