AI Innovation Scientist

Role : Junior Data Scientist

Experience : Minimum 4 years

Location : Hyderabad

About the role :

We are scaling an AI innovation team focused on practical, high-impact use cases for pharma manufacturers and commercial operations (CMO, CDMO, CRM). In this role you will rapidly prototype ML / AI solutions, work with cross-functional stakeholders (R&D, Process Development, Quality, Manufacturing, Commercial), and contribute to moving validated models toward production. This role is ideal for a pragmatic scientist / engineer with 4–5 years of applied data science experience who wants to focus on pharma problems (process analytics, PAT, sensor data, imaging, customer analytics) and learn regulated deployment practices.

Responsibilities :

➢ Collaborate with domain teams to translate business / regulatory problems into data science hypotheses and testable experiments.

➢ Acquire, clean, and integrate manufacturing (MES, LIMS, PAT, sensor / time-series, batch records), laboratory, and CRM / commercial datasets.

➢ Rapidly prototype models and algorithms : regression, tree ensembles, time-series forecasting, anomaly detection, clustering / segmentation, and basic deep learning (e.G., CNNs for imaging, RNNs / Temporal models).

➢ Build explainability and uncertainty estimates into prototype models for regulated decision-support.

• ➢ Validate model performance using cross-validation, holdout sets, and domain-appropriate metrics;

participate in documentation needed for audits / validation.

➢ Implement monitoring experiments (drift detection, simple retraining pipelines) and hand off monitoring requirements.

follow data governance and security policies.

translate model resultsinto actionable recommendations.

Required qualifications :

➢ 4–5 years of hands-on experience in applied data science, machine learning, or analytics (industry or research with applied projects).

Master’s preferred).

working knowledge of SQL.

➢ Practical experience with time-series / sensor data or tabular modeling in production-like settings.

➢ Experience with at least one deep learning framework (PyTorch or TensorFlow) for applied tasks. ➢ Demonstrated ability to move from problem definition to prototype and present results to stakeholders.

➢ Good documentation practices, basic testing, and reproducible analysis (notebooks + script refactors).

➢ Clear communication skills and ability to work in cross-functional teams.

➢ Willingness to work in regulated environments and follow documentation / validation processes.

Preferred qualifications :

➢ Prior experience in pharma, biotech, CMO / CDMO, manufacturing, or other regulated industries. ➢ Familiarity with MES, LIMS, ELN, PAT, or industrial IoT data sources.

➢ Experience with MLOps basics (Docker, simple CI / CD, MLflow, Airflow) or production handoffs. ➢ Knowledge of multivariate statistical process control (MSPC), DOE, chemometrics, or Six Sigma concepts.

➢ Exposure to LLMs and prompt engineering for knowledge extraction, summarization or augmentation of domain content (SOPs, batch records).

➢ Experience with cloud platforms (AWS / Azure / GCP) and data platforms (Snowflake, Redshift, BigQuery).

➢ Understanding of model explainability (SHAP, LIME) and uncertainty quantification techniques.

Technical stack :

➢ Languages : Python (pandas, scikit-learn, xgboost / lightgbm), SQL

➢ DL : PyTorch or TensorFlow / Keras

➢ Time-series : tsfresh, statsmodels, prophet, tslearn

➢ MLOps / infra : Docker, MLflow, Airflow (familiarity)

➢ Storage / Viz : S3 / object store, Postgres, Tableau / Power BI / Plotly ➢ Tools : Git, Jupyter / VS Code, basic Linux shell