Machine Learning Engineer

Domain: EV Battery Manufacturing & Electrochemistry

Experience: 5–7 Years

Education: B.Tech/M.Tech in Chemical Engineering or M.Sc. Chemistry

Role Objective

We are seeking a high-calibre Senior Machine Learning Engineer (L5) to bridge electrochemical research and gigafactory-scale manufacturing.

This role demands a rare combination of:

• Strong Chemistry / Chemical Engineering fundamentals
• Advanced Machine Learning & Deep Learning expertise
• Rigorous statistical and probabilistic thinking

You will build physics-informed digital twins to predict battery life, optimize manufacturing yield, and enable intelligent decision-making at scale.

Core Responsibilities

1. Advanced Machine Learning & Deep Learning

• Design and deploy time-series models (Transformers, LSTMs) to analyze battery cycling and degradation patterns
• Develop computer vision systems (CNNs, Vision Transformers) for defect detection in electrode coating and assembly
• Build physics-informed models (PINNs) embedding electrochemical constraints into learning frameworks
• Implement self-supervised and representation learning on large-scale industrial datasets

2. Generative AI & Intelligent Systems (good to have)

• Develop RAG-based systems to extract insights from chemical literature, patents, and technical documents
• Build agentic workflows / multi-agent systems for automated root-cause analysis across plant and lab data
• Enable knowledge-driven AI systems linking process, material, and performance data

3. Statistical Modelling & Scientific Rigor

• Lead Design of Experiments (DOE) for new materials and process optimization
• Apply multivariate statistical analysis (ANOVA, MANOVA) to understand process variability
• Develop probabilistic models (Gaussian Processes, Monte Carlo methods) for:
• Remaining Useful Life (RUL)
• Battery reliability and uncertainty quantification
• Implement statistical quality control (CUSUM, EWMA) for early drift detection

4. Physics-Informed & Domain-Driven Modelling

• Translate electrochemical principles into ML features and constraints
• Interpret EIS/Nyquist plots and integrate insights into predictive models
• Build hybrid models combining:
• First-principles physics
• Data-driven learning

5. Industrial AI & Deployment

• Integrate models with OT systems (SCADA/PLC) aligned with ISA-95 architecture
• Develop scalable pipelines using modern MLOps frameworks
• Deploy models for real-time decision support and optimization

Technical Skills

Category

Specific Technical Skills

Deep Learning

Physics‑Informed Neural Networks (PINNs), Transformers, Long Short‑Term Memory networks (LSTMs), Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs) for synthetic data generation, Autoencoders for anomaly detection

Machine Learning

Gradient Boosting (XGBoost, LightGBM), Random Forests, Support Vector Machines (SVMs), Clustering techniques (K‑Means, DBSCAN) for cell sorting and pattern discovery

Statistics

Bayesian Inference, Hypothesis Testing, P‑value analysis, Linear and Non‑linear Regression, Survival Analysis for longevity and reliability modeling

Mathematical Foundations

Linear Algebra (SVD, Eigen‑decomposition), Calculus (Gradients, Jacobians), Real Analysis, Optimization Theory

Machine Learning & AI

Transformers, LSTMs, CNNs, PINNs, Autoencoders, GANs, Gradient Boosting (XGBoost, LightGBM)

Programming & Platforms

Python, PyTorch, TensorFlow, MLflow, Docker, Kubernetes, Azure AI, Databricks

MLOps & Tools

Model lifecycle management, experiment tracking, containerization, scalable deployment using MLflow, Docker, Kubernetes, and Azure‑based data and AI platforms

Domain Requirements

• Strong foundation in:
• Electrochemistry
• Reaction kinetics
• Thermodynamics
• Understanding of battery systems (Li-ion preferred)
• Experience with industrial data environments and sensor systems