Lead Data Scientist

Sourcebae; Chennai; Presencial; INR 4.200.000 - 6.400.000; Jornada completa

Descripción de la vacante

Sourcebae seeks a Lead Data Scientist in Chennai to set direction for ML workstreams and lead a small team. You will guide architecture, manage end-to-end ML projects and serve as a technical authority for client discussions.

You will own the complete lifecycle, from data framing to monitoring, with emphasis on Snowflake, Spark, and modern MLOps practices across cloud platforms.

Formación

• 8+ years of Data Science experience.
• Strong hands-on expertise in classical ML including regression, classification, XGBoost and LightGBM.
• Strong deep learning experience with architecture selection and training at scale.
• Deep expertise in NLP, Forecasting, Recommendation Systems or Computer Vision.
• Genuine hands-on Snowflake experience, including architecture and data/training decisions.
• Production-scale feature engineering and ML pipelines.
• Distributed training using Spark, Ray or equivalent.
• Ownership of production ML models, including monitoring, drift detection, retraining and rollback.
• Strong technical leadership and mentoring experience.
• Excellent client-facing communication and ability to defend technical decisions.
• Hands-on experience with cloud ML platforms such as Azure ML, Databricks, SageMaker or Vertex AI.
• GenAI/LLM experience beyond demo/prototype projects.
• Strong MLOps experience with MLflow or similar tools, automated retraining and ML CI/CD.
• Consulting/professional services experience managing multiple client engagements.

Responsabilidades

• Own the complete model lifecycle from problem framing and feature engineering through training, validation, deployment, monitoring and retraining.
• Defend model architectures, including classical ML vs deep learning decisions.
• Design experiments including A/B testing, causal inference and uplift modeling.
• Build production-grade feature pipelines and training infrastructure.
• Work with distributed computing frameworks such as Spark, Ray or equivalent.
• Monitor production models for drift and degradation and determine retraining strategies.
• Set technical standards for modeling, code reviews, experiment tracking and validation.
• Mentor and guide 2-4 Data Scientists/Engineers.
• Act as the primary technical contact for clients.
• Explain model trade-offs and assumptions to non-technical stakeholders.
• Drive GenAI/LLM solutions including retrieval design and evaluation frameworks.
• Implement MLOps practices with model registries, automated retraining and ML CI/CD.

Conocimientos

XGBoost

LightGBM

Python

SQL

Snowflake

Databricks

SageMaker

Vertex AI

Spark

Ray

GenAI/LLM

A/B testing

MLOps

Model Monitoring

Client-facing

Herramientas

Snowflake

Databricks

SageMaker

Vertex AI

Spark

Ray

MLflow

Azure ML

Descripción del empleo

Required Skills

• Classical ML – XGBoost, LightGBM, Regression, Classification
• Snowflake – Mandatory, genuine hands-on platform experience
• Python & SQL
• Databricks / SageMaker / Vertex AI or equivalent
• Spark / Ray or equivalent distributed computing
• GenAI / LLM – RAG, evaluation, monitoring and failure-mode analysis

Role Overview

We are looking for a Lead Data Scientist who can set the technical direction for a workstream, make architecture and modeling decisions, and lead a small team of Data Scientists/Engineers.

The candidate should have strong hands-on experience across the complete ML lifecycle and be capable of acting as the senior technical authority for client-facing discussions.

Key Responsibilities

• Own the complete model lifecycle from problem framing and feature engineering through training, validation, deployment, monitoring and retraining.
• Select and defend appropriate model architectures, including decisions between classical ML and deep learning.
• Design and execute rigorous experiments including A/B testing, causal inference and uplift modeling.
• Build production-grade feature pipelines and training infrastructure.
• Work with distributed computing frameworks such as Spark, Ray or equivalent.
• Monitor production models for drift and degradation and determine retraining, redesign or rollback strategies.
• Establish technical standards for modeling, code reviews, experiment tracking and model validation.
• Mentor and technically guide 2-4 Data Scientists/Engineers.
• Act as the primary technical point of contact for clients.
• Explain model trade-offs, limitations, assumptions and failure modes to non-technical stakeholders.
• Drive GenAI/LLM solutions including retrieval design, evaluation frameworks and production reliability.
• Implement MLOps practices including model registries, automated retraining and ML CI/CD.

Must-Have Requirements

• 8+ years of Data Science experience.
• Strong hands-on expertise in classical ML including regression, classification, XGBoost and LightGBM.
• Strong deep learning experience with architecture selection and training at scale.
• Deep expertise in at least one area: NLP, Forecasting, Recommendation Systems or Computer Vision.
• Genuine hands-on Snowflake experience, including architecture and data/training decisions.
• Production-scale feature engineering and ML pipelines.
• Distributed training / large-scale compute experience using Spark, Ray or equivalent.
• Proven ownership of production ML models, including monitoring, drift detection, retraining and rollback.
• Strong technical leadership and mentoring experience.
• Excellent client-facing communication and ability to defend technical decisions.
• Hands-on experience with cloud ML platforms such as Azure ML, Databricks, SageMaker or Vertex AI.
• Practical GenAI/LLM experience beyond demo/prototype projects.
• Strong MLOps experience with MLflow or similar tools, automated retraining and ML CI/CD.
• Consulting/professional services experience with the ability to manage multiple client engagements.