Jd Role Descriptions: Key ResponsibilitiesDesign| develop| and own predictive AIML models for complex systems (e.g.| forecasting| degradation| pattern recognition).Build and evaluate deep learning models| including CNNs| where appropriate to the problem domain.Apply model explainability techniques to interpret black box behaviour and communicate insights to technical and business stakeholders.Develop hybrid models that integrate empirical rules| domain knowledge| or physics informed assumptions with ML techniques.Perform rigorous model validation| robustness testing| and error analysis under real world data constraints.Translate ambiguous business or engineering problems into well defined AIML problem statements with transparent assumptions and success criteria.Collaborate with domain experts| data engineers| and platform teams to ensure models are trustworthy| usable| and production ready.Document model logic| assumptions| limitations| and interpretability clearly for governance and long term sustainability.
Essential Skills:
Required Skills Experience Core58 years of hands on experience in AIML or Data Science roles with strong focus on predictive modelling.Proven experience with machine learning and deep learning| including CNNs.Strong understanding of model behaviour| uncertainty| bias| and explainability.Demonstrated experience building hybrid AIML models (empirical data driven or physics informed ML).Strong proficiency in Python and AIML libraries.Solid grounding in statistics| applied mathematics| and modelling fundamentals.Ability to clearly explain complex AIML models and trade offs to stakeholders.AI ML Stack (Indicative)Python based AIML and deep learning frameworksCNN architectures and predictive modelling techniquesModel interpretability and validation methodsSQL and data platforms for large scale model training and evaluationWhat We Look ForStrong AIML intuition beyond off the shelf models.Ability to work in the grey zone between empirical understanding and data driven AI.Ownership mindset with focus on model credibility and impact.Comfort explaining and defending model choices and limitations.