ROLE OVERVIEW

We are looking for a Data Scientist with 3+ years of strong hands-on experience in Machine Learning, Deep Learning, Databricks, and production-oriented AI/ML solution development. The ideal candidate should be capable of independently owning model experimentation, feature engineering, model training, evaluation, tuning, deployment support, monitoring, and optimization for real-world business use cases. The role requires strong working expertise in Databricks, Spark, PySpark, Delta Lake, MLflow, Python-based ML frameworks, and practical MLOps practices, with only working-level exposure to Azure cloud services used for data and AI workloads.

KEY RESPONSIBILITIES

- Independently design, develop, validate, optimize, and improve Machine Learning models for classification, regression, clustering, forecasting, recommendation, anomaly detection, ranking, NLP, and predictive analytics use cases.
- Build and tune Deep Learning models using neural networks, CNNs, RNNs, LSTMs, transformers, autoencoders, embeddings, transfer learning, and fine-tuning techniques where applicable.
- Develop scalable Databricks-based workflows using notebooks, jobs, workflows, clusters, Delta Lake, Spark, PySpark, MLflow, experiment tracking, model registry, and reusable pipeline components.
- Perform advanced feature engineering, statistical analysis, model selection, cross-validation, hyperparameter optimization, error analysis, explainability assessment, and performance benchmarking.
- Work with structured, semi-structured, and unstructured data from lakehouse platforms, enterprise systems, APIs, documents, text, images, logs, and time-series datasets.
- Apply NLP and text analytics techniques including embeddings, semantic search, text classification, entity extraction, summarization, sentiment analysis, similarity matching, and transformer-based model usage.
- Support production MLOps activities including model packaging, deployment support, monitoring, drift analysis, retraining strategy, experiment governance, model versioning, and operational documentation.
- Collaborate with data engineers, senior data scientists, architects, and business stakeholders to translate complex business problems into measurable ML and Deep Learning solutions.
- Own model documentation, experiment comparison, assumptions, evaluation metrics, limitations, risks, recommendations, and stakeholder-ready technical summaries.

REQUIRED EXPERIENCE

- 3–4 years of experience in Data Science, Machine Learning, Deep Learning, AI engineering, advanced analytics, or related roles.
- Strong hands-on experience building, tuning, validating, and deploying ML models using Python and libraries such as Scikit-learn, XGBoost, LightGBM, Pandas, NumPy, SciPy, Statsmodels, TensorFlow,



PyTorch, Keras, and MLflow.
- Solid practical experience with Deep Learning architectures such as feed-forward neural networks, CNNs, RNNs, LSTMs, transformers, autoencoders, embeddings, attention mechanisms, transfer learning, and model fine-tuning.
- Strong understanding of the full ML lifecycle including data preparation, feature engineering, training, validation, tuning, testing, packaging, deployment support, monitoring, drift detection, retraining, and continuous improvement.
- Strong hands-on experience with Databricks for large-scale data processing, feature engineering, model experimentation, distributed compute, MLflow tracking, model registry, and production pipeline execution.
- Strong experience with Spark, PySpark, Delta Lake, SQL, distributed processing, lakehouse patterns, scalable data pipelines, and performance optimization of data workflows.
- Working exposure to Azure services such as Azure Data Lake, Azure Machine Learning, Azure OpenAI, Azure Data Factory, Azure DevOps, model endpoints, and deployment workflows is desirable but secondary to Databricks and ML depth.
- Exposure to Generative AI concepts such as LLMs, prompt engineering, embeddings, RAG architecture, vector search, evaluation, guardrails, and responsible AI practices is preferred.
- Understanding of Agentic AI concepts such as AI agents, orchestration, tool usage, workflow automation, task planning, and multi-step reasoning is an added advantage.
- Ability to explain model performance, evaluation metrics, trade-offs, limitations, failure patterns, risks, and recommendations clearly to business and technical stakeholders.

TECHNICAL SKILLS

Skill Area

Expected Expertise

Databricks

Advanced use of Databricks notebooks, Jobs, Workflows, clusters, Delta Lake, Spark, PySpark, MLflow, experiment tracking, model registry, feature engineering, scalable pipeline execution, and performance tuning

Machine Learning

Classification, regression, clustering, forecasting, recommendation, anomaly detection, ensemble models, feature engineering, hyperparameter tuning, validation strategy, explainability, benchmarking

Deep Learning

Neural networks, CNNs, RNNs, LSTMs, transformers, autoencoders, embeddings, attention mechanisms, transfer learning, model fine-tuning, training optimization, GPU-based training concepts

Python ML Stack

Python, Pandas, NumPy,



Scikit-learn, XGBoost, LightGBM, TensorFlow, PyTorch, Keras, Statsmodels, SciPy, feature engineering utilities, reusable notebook and package development

NLP and Text Analytics

Text classification, semantic search, entity extraction, summarization, sentiment analysis, similarity matching, embeddings, transformer-based NLP models, document intelligence use cases

MLOps

MLflow, experiment tracking, model registry, model versioning, reproducible pipelines, model packaging, deployment support, monitoring, drift detection, retraining workflows, CI/CD awareness

Azure Basics

Working exposure to Azure Data Lake, Azure Machine Learning, Azure OpenAI, Azure Data Factory, Azure DevOps, model endpoints, secrets, storage, and deployment workflows

Generative AI

LLMs, prompt engineering, RAG, embeddings, vector search, evaluation basics, guardrails, AI application design, responsible AI awareness

Programming and Data

Python, SQL, PySpark, REST APIs, Git, reusable code development, data validation, pipeline troubleshooting, performance optimization, documentation

Visualization

Power BI, dashboards, model performance reporting, KPI reporting, data storytelling, stakeholder-ready insights, experiment comparison summaries

PREFERRED QUALIFICATIONS

- Bachelor's or Master's degree in Computer Science, Data Science, Statistics, Mathematics, Engineering, Artificial Intelligence, Machine Learning, or a related discipline.
- Relevant certifications in Databricks, Data Science, Machine Learning, Deep Learning, MLOps, or Generative AI are preferred.
- Experience with Databricks-based ML workflows, feature engineering, scalable experimentation, model evaluation, MLflow tracking, model registry, and reusable ML components.
- Exposure to deep learning model development, NLP use cases, embeddings, model governance, production support, and model monitoring is preferred.
- Ability to work in agile delivery environments and contribute to demos, documentation, experimentation reports, model comparison summaries, risk notes, and stakeholder reviews.

EXPECTED OUTCOMES

- Deliver production-ready Machine Learning and Deep Learning models using Databricks-led development and scalable data processing practices.
- Improve business decision-making through robust predictive models, advanced analytics, NLP solutions, model explainability, and reusable AI/ML components.
- Build repeatable model experimentation workflows, feature engineering pipelines, evaluation reports, MLflow-tracked assets, model versioning, and monitoring-ready artifacts.
- Support innovation across Machine Learning, Deep Learning, Generative AI, and Databricks-based production AI implementation patterns.

📌 Data Scientist (Mumbai)
🏢 Exponentia.ai
📍 Mumbai