Staff Engineer - Machine Learning

The Eplane Company; Chennai; On-site; INR 900,000 - 1,500,000; Full time

Job summary

The Eplane Company is seeking a seasoned engineer to build models and tooling that leverage optimization, multi-physics models, and machine learning for engineering, operations, and commercial teams. You will help transform mathematical models, aerospace physics simulations, and reinforcement learning into GPU-accelerated software engines for high-throughput, low-latency use.

You will lay the architectural and algorithmic foundation for our internal design optimisation and operations suite,

Qualifications

• Expertise in advanced autonomous systems where optimization, ML, and software architecture meet.
• Strong foundation in mathematics: linear algebra, optimization, ML, probability, differential geometry.
• Deep understanding of mathematical optimization from gradient-based to non-gradient methods.
• Proficiency in ML frameworks (PyTorch, JAX) and applying ML to physical/simulated environments.
• Knowledge of Markov Decision Processes and POMDP solvers.

Responsibilities

• Design the optimization and representation architecture of our platform across internal engineering tools and aircraft operations software.
• Build and validate coupled ML models and systems across complex multi-physics systems, planning and optimisation systems.
• Develop tooling to evaluate trade-offs between aircraft parameters and fleet-level metrics.
• Architect C++ and Python APIs that connect physics solvers to user-facing tools and simulators.
• Own verification and validation, and advise Engineering on model trust.

Skills

Optimization

Machine Learning

C++

Python

PyTorch

JAX

Algorithm design

Geospatial indexing

MDP/POMDP

High-performance software

Tools

PyTorch

JAX

Job description

Job Summary

We are looking for a candidate who can build models and tooling leveraging optimisation, multi-physics models, and machine learning to be used across our engineering, operations, and commercial teams. You will join our team to transform mathematical models, aerospace physics simulations, and Reinforcement Learning algorithms into high-throughput, low-latency, production-grade GPU-accelerated software engines. This role will lay the the architectural and algorithmic foundation for our internal engineering design optimisation and operations suite.

Responsibilities

• Design the optimisation and representation architecture of our platform across both internal engineering tools and aircraft operations software
• Build and validate coupled ML models and systems across complex multi-physics systems, planning and optimisation systems
• Develop the algorithmic tooling to evaluate trade-offs between aircraft parameters parameters (e.g., rotor configuration, battery size) and fleet-level metrics (e.g., vertiport throughput, community noise impact).
• Architect the C++ and Python APIs that connect the underlying physics solvers to user-facing engineering design tools and macro-level network simulators
• Own verification and validation, and tell the Engineering teams on where a model can and cant be trusted.

Requirements

• Required Qualifications Expertise in advanced autonomous systems where optimization, Machine Learning, and software architecture meet
• Understanding of mathematics, particularly linear algebra, optimisation, Machine Learning, probability theory, differential geometry
• Deep understanding of mathematical optimization across gradient based optimisation to non-gradient based optimization
• Strong practical experience with ML frameworks (PyTorch, JAX) and applying ML to physical or simulated environments
• Understanding of Markov Decision Processes and methods to solve POMDPs
• Exceptional proficiency in modern C++ and Python. You know how to build maintainable, high-performance software systems.
• Prior experience in architecture design for data/ML systems; defining success metrics, delivery plans and quality
• Experience with 3D or other irregular data: meshes, point clouds, voxels, sparse tensors.
• Proven ability to implement and optimize high-throughput spatial queries using R-trees, Bounding Volume Hierarchies, KD-trees, or geospatial indexing systems

Preferred Qualifications

• Familiarity with surrogate modelling using system identification, Gaussian processes, neural networks
• Expertise with Automatic differentiation tools and development
• Deep knowledge of MDO frameworks and aircraft sizing software.
• Understanding of hybrid-electric propulsion integration at the aircraft level