ML Engineer

About Gradera

Gradera defines a new category of enterprise transformation called Software-Orchestrated Services™ - where software orchestrates human expertise, digital workers, and enterprise systems to deliver governed outcomes at scale. As an AI Native Services firm, we help enterprises redesign how work gets done across operations, product, engineering, customer experience, data, and enterprise workflows to move beyond fragmented AI pilots and disconnected automation toward measurable business outcomes.

Overview

We are seeking skilled ML Engineers to join our Simulation & Scenario Enablement team. This is a specialized role at the intersection of machine learning engineering and physics-based simulation. You will design and implement production-grade ML pipelines, build physics-informed neural networks (PINNs) that respect physical constraints, and develop neural architectures that accelerate simulation workloads. You will own the full MLOps lifecycle — from feature engineering and model training to deployment, monitoring, and continuous improvement — ensuring ML models reliably power real-time scenario evaluation and digital twin intelligence.

Our core ML engineering stack includes:

ML Frameworks & Development

• PyTorch and TensorFlow for neural network development
• Physics-Informed Neural Networks (PINNs) for constraint-aware modeling
• Neural ODE solvers (torchdiffeq, diffrax) for continuous-time dynamics
• Python (NumPy, SciPy, pandas) for numerical computing
  
  

MLOps & Platform

• Databricks ML for scalable model training and pipelines
• MLflow for experiment tracking, model registry, and deployment
• Unity Catalog for ML asset governance and lineage
• Delta Lake for feature storage and versioned training data
• Feature Store for feature management and serving
  
  

Production & Monitoring

• Model serving and inference optimization
• Model monitoring, drift detection, and alerting
• CI/CD for ML pipelines
• Containerized model deployment (Docker, Kubernetes/OpenShift)
  
  

Key Responsibilities

• Design and implement Physics-Informed Neural Networks (PINNs) with domain constraints
• Develop neural ODE solvers and surrogate models for physics simulations
• Build hybrid ML architectures combining data-driven learning with physics-based priors
• Optimize neural models for accuracy, inference speed, and resource efficiency
• Design scalable feature engineering pipelines using Databricks and PySpark
• Manage features in Feature Store and build Delta Lake training pipelines
• Build end-to-end ML pipelines on Databricks ML
• Track experiments, version models, and deploy using MLflow
• Implement model monitoring for drift, performance, and prediction quality
• Build CI/CD for ML and ensure governance via Unity Catalog
  
  

Preferred Qualifications

• 7+ years of experience in ML engineering, applied ML, or scientific computing roles
• Master’s or PhD in Computer Science, Machine Learning, Computational Science, Physics, or related field
• Track record of deploying ML models in production at scale
• Experience with physics-based or scientific ML applications
• Experience working in agile, cross-functional teams
  
  

Highly Desirable

• Experience with ML for digital twin or simulation platforms
• Background in computational physics, numerical methods, or scientific computing
• Experience with differentiable programming and automatic differentiation frameworks
• Familiarity with discrete event simulation or agent-based modeling integration
• Experience with GPU-accelerated training and inference optimization
• Publications or patents in physics-informed ML, neural ODEs, or surrogate modeling
• Contributions to open-source ML/scientific computing projects
• Exposure to industrial domains such as Manufacturing, Logistics, or Transportation is a plus