Engineer II - Robotics MLOps

NewSpace Research and Technologies; Bengaluru; On-site; INR 1,400,000 - 2,100,000; Full time

Job summary

NewSpace Research and Technologies in Bengaluru is seeking an Engineer II - Robotics MLOps to bridge validated R&D algorithms with field deployment. You will manage the release lifecycle, ensure hardware-aware ML optimization, and coordinate with R&D and DevOps teams for reliable flight-test deployments.

You will champion CI/CD, model versioning, and rigorous testing in a fast-paced robotics environment, balancing real-time constraints with software quality and safety requirements.

Qualifications

• Bachelor’s or Master’s degree in ML/CS/Embedded/Electrical/Mechatronics.
• 3+ years (Bachelor’s) or 1+ years (Master’s) of industrial ML/robotics deployment.
• Deep proficiency in neural networks and hardware-specific optimization (TensorRT, CUDA).
• Strong C++ and Python in Linux-based robotics environments.
• Experience with TDD and automated testing (GTest, PyTest).
• CI/CD with build automation and Git-based branching.
• Profiling and debugging embedded software (Nsight, GDB, Valgrind).
• Docker, Kubernetes (K3S) in embedded environments.

Responsibilities

• Release Lifecycle Management: manage gated promotion of software across repos and flight-test alignment.
• ML Deployment & Optimization: optimize models for embedded deployment using quantization, pruning, distillation; tune with CUDA/TensorRT.
• CI/CD & MLOps Infrastructure: design and maintain the MLOps toolchain with automated training and versioning.
• Release Readiness: lead release candidate prep; coordinate logs and performance reports for stakeholders.
• Observability & Diagnostics: implement logging and health monitoring for flight-test data.
• Infrastructure & Pipeline Reliability: maintain dev environments, containers, and deployment toolchains.

Skills

ML deployment

Neural nets

CUDA

TensorRT

C++

Python

Linux

GTest

PyTest

Nsight

GDB

Valgrind

Docker

Kubernetes (K3S)

Git/BitBucket

Education

Bachelor's degree in ML/CS/Engineering

Master’s degree in ML/CS/Engineering

Tools

Docker

Kubernetes (K3S)

Nsight

GDB

Valgrind

Git

BitBucket

Job description

Denied Operations Division > GNSS-Denied

Domain:

Robotic Systems

What this Role Offers

• Full-Stack ML Deployment: Own the deployment of advanced robotics and ML algorithms onto high-performance edge hardware
• Hardware Acceleration Ownership: Direct hands-on work with NVIDIA platforms, utilizing CUDA and TensorRT to squeeze real-time performance out of complex neural networks
• Gatekeeper of Software Releases: Act as the critical final validation point in a gated release process, ensuring only the most performant, stable code reaches our UAV fleets
• Impactful Technical Intersection: Operate at the nexus of MLOps, DevOps, and Robotics Engineering, where you ensure that cutting-edge R&D algorithms are optimized for mission-critical reliability
• Systematic Release Governance: The authority is held to define and enforce quality gates, ensuring that only verified, performance-tuned robotics code is promoted to mission-critical deployment branches

About the Role

The software release and deployment lifecycle is architected by the Engineer II - Robotics MLOps. A critical bridge is maintained between validated R&D algorithms and field-ready deployment. This role ensures the stability of the branching strategy, manages the gated promotion of code from Staging to Development and Release, and provides the architecture for MLOps and DevOps pipelines. This position is designed for an engineer, by whom it is understood that ML in robotics is not just about the model - it is about the hardware constraints, real-time performance, and the rigour of the release process

Key Responsibilities

• Release Lifecycle Management: Manage the gated promotion of software across repositories; coordinate the movement of code from Staging to Development and eventually to Release branches, aligning with flight-test verification and QA standards
• ML Deployment & Optimization: Optimize machine learning models and robotics algorithms for embedded deployment by applying advanced techniques such as quantization, pruning, and knowledge distillation. Profile and tune models for hardware acceleration using CUDA, TensorRT, and other NVIDIA platform primitives
• CI/CD & MLOps Infrastructure: Design and maintain the MLOps toolchain, including automated training, model versioning, and deployment pipelines that are validated against flight-test data
• Release Readiness: Act as the technical lead for release candidate preparation. Coordinate with Senior Systems Engineers to ensure all logs, telemetry, and performance reports are compiled for final stakeholder approval
• Observability & Diagnostics: Implement logging and health monitoring infrastructure to ensure flight-test data is structured, accessible, and actionable for the R&D team
• Infrastructure & Pipeline Reliability: Maintain development environments, containers, and deployment toolchains, serving as a secondary point of contact for R&D, Embedded, and DevOps workflows

Minimum Qualifications

• Bachelor's or Master’s degree in Machine Learning, Computer Science, Embedded Systems, Electronics, Electrical Engineering, or Mechatronics
• 3+ years (Bachelor’s) or 1+ years (Master’s) of industrial experience in ML and/or robotics software deployment
• ML Production Skills: Deep proficiency in Neural Networks and hardware-specific model optimization (TensorRT, CUDA)
• Software Rigour: Strong proficiency in C++ and Python with a focus on system performance, memory management, and multi-threading in Linux-based robotics environments
• Testing Standards: Demonstrated experience in Test-Driven Development (TDD) and the use of automated testing frameworks (e.g., GTest, PyTest)
• CI/CD Expertise: Proven experience with build automation, artefact versioning, and complex branching strategies (BitBucket/Git)
• Profiling & Debugging: Experience in profiling and debugging embedded software using tools like Nsight, GDB, or Valgrind
• Deployment Tooling: Experience with Docker, Kubernetes (K3S), or similar containerization tools in embedded environments

Preferred Qualifications

• Domain Expertise: Familiarity with SLAM, NeuRF and Gaussian Splatting, Sensor Fusion, or Computer Vision pipelines for navigation
• Safety Standards: Experience with safety-critical software development standards (e.g., MISRA C++)
• Hardware Platforms: Hands-on experience with specific NVIDIA edge platforms (Jetson Orin/Xavier)
• Real-Time Systems: Experience with RTOS or real-time Linux kernel customization
• Evaluation Pipelines: Experience in designing and managing data-driven evaluation pipelines for robotics/ML performance metrics

Working Hours

• Standard working hours are 9:30 AM to 6:30 PM, Monday to Friday
• Field-testing activities may require early-morning or extended hours, depending on mission requirements

Compensation Range

• Competitive compensation aligned with industry standards, including performance-based incentives
• Exact salary ranges will be customised according to experience
• Detachment allowance

Required Skills

DevOps MLOps Machine learning : OpenCV, TensorFlow or PyTorch