Data Scientist

The Applied ML Engineer will design, build, and operationalize machine-learning models that power content production, localization, metadata enrichment, archival workflows, and intelligent search/retrieval across large-scale media systems. This role sits at the intersection of applied machine learning, content intelligence, and production-grade engineering-supporting data-driven decisions and automation across the content supply chain., * Develop, train, and optimize models for media metadata extraction, content classification, entity resolution, similarity search, and multimodal understanding.

• Build predictive and prescriptive models to streamline content operations such as localization quality prediction, asset matching, retrieval ranking, and automated tagging.
• Conduct rigorous analysis, feature engineering, and model selection using modern statistical and ML frameworks.

2. Production-Grade ML Engineering

• Implement scalable ML pipelines using Python, cloud-native services, and enterprise data platforms.
• Partner with Data Engineering teams to design performant data flows for model training, validation, and inference across high-volume media catalogs.
• Build robust evaluation frameworks and monitoring systems ensuring quality, reliability, and drift detection in production environments.

3. MLOps & Model Deployment

• Containerize, deploy, and maintain ML services using CI/CD, orchestration frameworks, and real-time or batch inference architectures.
• Collaborate with platform and infrastructure teams to integrate models with content production systems, search platforms, APIs, and metadata services.
• Ensure reproducibility, versioning, and lifecycle management aligned with enterprise machine-learning practices.

4. Media Domain Expertise (Nice to have)

• Apply ML techniques to domain-specific challenges in:

• Content production: post-production signals, QC automation, time-coded metadata, and asset lineage.

• Localization: subtitle/CC alignment, translation quality scoring, automated language metadata enrichment.

• Distribution formats: asset matching, technical metadata extraction, content packaging intelligence.

• Archival & retrieval: semantic search, embeddings, similarity models, knowledge graph augmentation.

• Work closely with media pipeline, operations, and creative engineering teams to ensure solutions align to real-world workflows.

5. Cross-Functional Collaboration & Stakeholder Engagement

• Partner with product managers, content operations, engineering teams, and metadata specialists to translate business needs into ML-driven solutions.
• Communicate complex model behavior, trade-offs, and results to technical and non-technical stakeholders.
• Contribute to solution roadmaps and technology evaluations for emerging ML techniques relevant to content intelligence.

6. Continuous Improvement & Innovation

• Stay current on advances in machine learning, multimodal modeling (text/audio/video), vector search, and media AI.
• Drive experimentation around next-generation retrieval models, embeddings, fine-tuning pipelines, and automated metadata generation.
• Evaluate and integrate third-party tools, open-source libraries, and cloud-native AI services to accelerate delivery.

Skills : Python Machine Learning, data science, AWS, Statistical Modeling, Semantic Search, Vector DB, GenAI, SQL, * Strong proficiency in Python, applied ML, and statistical modeling.

• Practical experience with media metadata, content understanding, search/retrieval, or multimodal ML.
• Hands-on background in MLOps, model deployment, and operationalizing ML workflows.
• Experience working in production-gradeenvironments with large-scale datasets and distributed systems.
• Proven ability to collaborate across engineering, operations, and product teams with clear, concise communication.