Data Scientist

Testunity; Pune District; On-site; INR 1,523,000 - 1,825,000; Full time

Job summary

Testunity is looking for a Data Scientist to design, train, and operationalize ML models with a strong engineering background. You will handle feature engineering, model training, inference, and post-processing on a cloud-native platform built on GCP and Kubernetes, emphasizing MLOps, reproducibility, and production readiness.

Onsite work is required across multiple Indian cities; bench candidates will contribute to end-to-end ML development, monitoring, and governance in production environments.

Qualifications

• Experience with Python for data science and ML use-cases.
• Hands-on with ML frameworks (e.g., PyTorch, scikit-learn, catboost).
• Familiarity with Kubernetes and Argo Workflows for ML pipelines.

Responsibilities

• Develop, train, and evaluate ML models in Python.
• Orchestrate ML workflows using Argo on Kubernetes.
• Leverage Vertex AI for scalable training and validation.
• Build runtimes for training and inference in Python.
• Monitor and optimize ML pipelines in production.

Skills

Python

Kubernetes

GCP

Communication

CI/CD

Data Science

PyTorch

scikit-learn

CatBoost

Tools

Argo Workflows

Vertex AI

GCS

Job description

Data Scientist (Bench candidate only)

Experience: 6-8 Yrs

Budget: 1.5LPM

UAN, BGV and PF history required

Onsite:

• Gurugram, Bengaluru, Pune, Jaipur, New Delhi, Noida, Hyderabad, Chennai, Bhopal, Mumbai / Navi Mumbai

Job Description

Role Overview

We are seeking a Data Scientist with a strong engineering background to design, build, train, and operationalize machine learning models that deliver measurable business impact. You will work end-to-end across feature engineering, model training, inference, and post-processing, leveraging a modern, cloud-native ML platform built on GCP and Kubernetes. This role blends strong statistical and machine learning expertise with hands-on MLOps practices, ensuring models are reliable, scalable, and production-ready.

Key Responsibilities

Model Development & Data Science

• Develop, train, and validate machine learning models using Python.
• Perform feature engineering, exploratory data analysis, and model evaluation.
• Apply appropriate ML techniques for prediction, classification, or optimization use cases.

ML Workflow Orchestration

• Use Argo Workflows on Kubernetes to orchestrate model inference and post-processing pipelines.
• Design repeatable, automated workflows for ML experiments and production inference.

Model Training & Validation

• Leverage Vertex AI to run scalable model training, hyperparameter tuning, and validation.
• Ensure reproducibility and consistency across training runs.

Model Runtime & Inference

• Build and maintain Python-based runtimes for training, inference, and feature engineering.
• Optimize inference pipelines for performance, reliability, and scalability.
• Handle and monitor pipelines in production environments.
• Participate in Oncall rotation for inference infrastructure.

Model Storage & Lifecycle Management

• Handle trained artifacts in Google Cloud Storage (GCS).
• Track model metadata, versions, and lineage using Argo or custom model registries.
• Support model versioning, rollback, and auditability.

Quality, Monitoring & Governance

• Define model evaluation metrics and validation criteria.
• Support post-deployment monitoring, drift detection, and retraining strategies.
• Follow best practices for documentation, testing, and responsible AI usage.

Required Skills & Qualifications

Technical Skills

• Strong on-hands proficiency in Python for data science and machine learning use-cases.
• Hands-on experience with ML frameworks (e.g., PyTorch, scikit-learn, catboost, ...).
• Familiarity with Kubernetes and Kubernetes-based workflows, especially Argo Workflows.

Data Science & ML Concepts

• Strong understanding of feature engineering, model evaluation, and validation techniques.
• Experience taking models from experimentation to production inference.
• Understanding of ML lifecycle management and MLOps principles.

Soft Skills

• Strong analytical and problem-solving mindset.
• Ability to translate business problems into data science solutions.
• Clear communication skills to explain models and results to diverse stakeholders.

Nice to Have

• Experience with real-time or batch inference systems.
• Exposure to CI/CD for ML pipelines.
• Familiarity with model monitoring, drift detection, and retraining strategies.

Skills: gcp,kubernetes,python,communication,ci/cd,data science