Machine Learning Developer | Sutherland | Chennai

Sutherland Global Services is seeking a Machine Learning Developer with 4+ years of experience in building high-performing, scalable, enterprise-grade ML systems. As an ML Developer, you'll work across the full machine learning lifecycle for the AI components of an existing platform: data collection and preparation, model selection and training, deployment, monitoring, and retraining as data drifts. You'll design and integrate NLP and machine translation components, build the data pipelines that feed them, and collaborate across developers, testers, and subject matter experts to translate business needs into production ML solutions. You'll work across both the setup phase and ongoing support, serving as the focal point for any NLP-driven and GenAI-driven features in the platform.

Responsibilities:

• Contribute across the full ML lifecycle including data collection, cleaning, preprocessing, model training, deployment, and post-release monitoring.
• Design and implement machine learning models using appropriate algorithms for natural language processing, machine translation, and document understanding tasks.
• Build and maintain the data pipelines that feed the models, including ingestion, transformation, and feature preparation.
• Evaluate model performance using established metrics, monitor for data drift, and retrain or adjust models as production data evolves.
• Integrate and tune machine translation quality estimation models such as COMET, COMETKiwi, OpenKiwi, or TransQuest to provide segment-level confidence without human reference translations.
• Integrate adaptive machine translation systems such as ModernMT or equivalent that learn from human post-editor corrections in real time, alongside general neural machine translation providers such as Azure OpenAI, Azure Translator, Google Translate, and DeepL.
• Apply automatic post-editing techniques and large language model prompting to correct or improve machine translation output where appropriate.
• Deploy ML models and services into production on Azure and monitor them post-release.
• Apply sound software engineering practices including object-oriented design, unit-test-driven development, and clear architecture to ML code.
• Stay updated with the latest advancements in machine learning, NLP, machine translation, and GenAI, and contribute to the team's continuous improvement.
• Collaborate with cross-functional teams including developers, testers, and subject matter experts to translate business requirements into scalable and efficient ML solutions.

Qualifications:

• Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Data Science, Computational Linguistics, or a related field.
• 4+ years of experience building production machine learning systems end-to-end, from data preparation through deployment and monitoring.
• Strong proficiency in Python and familiarity with mainstream ML libraries and frameworks (TensorFlow, PyTorch, scikit-learn, pandas, NumPy, spaCy, NLTK).
• Comfort with Jupyter notebooks for experimentation and with regular expressions for text processing.
• Hands-on experience with NLP techniques and applications, including tokenization, embeddings, attention, sentence splitting, and language modeling.
• Working understanding of GenAI applications and large language model prompting for production use cases.
• Experience building and maintaining data pipelines using Databricks, Airflow, or equivalent.
• Familiarity with machine translation evaluation and quality estimation, including metrics such as BLEU, chrF, METEOR, COMET, and COMETKiwi.
• Familiarity with adaptive machine translation, transformer-based model fine-tuning, glossary enforcement, and human-in-the-loop workflows.
• Sound software engineering fundamentals including object-oriented design, unit-test-driven development, and architectural judgment.
• Strong grasp of data structures, data modeling, and SQL for data manipulation and analysis.
• Strong communication skills and the ability to explain ML concepts clearly to non-technical stakeholders including subject matter experts and project managers.
• Strong analytical and problem-solving ability and a collaborative working style across developers, testers, and SMEs.
• Experience with Java or C/C++ as a secondary language, automatic post-editing, PyMuPDF or Azure Document Intelligence for document processing, familiarity with localization tools such as Trados, memoQ, XTM, or Phrase, or prior production work on machine translation or virtual agent platforms would be advantageous.