Sr. Engineer III - CVML

NewSpace Research and Technologies; Bengaluru; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

NewSpace Research and Technologies seeks a Senior Computer Vision & Machine Learning Engineer to architect and deliver learning-based visual navigation and resilient-PNT capabilities. You will own direction for learned features, day-night adaptation, camera-IMU integration and embedded ML optimization, coordinating across perception, state estimation and flight systems.

You will mentor engineers, drive system-level validation, and lead cross-functional decisions while advancing multimodal

Qualifications

• Bachelor’s, Master’s or PhD in Robotics/EE/CS or related field as minimum.
• Strong Python, C++, and ROS proficiency.
• Advanced practical experience with modern deep-learning frameworks and computer-vision libraries.
• Experience building visual localization such as visual-inertial odometry, SLAM, scene matching or related navigation systems.
• Deep understanding of camera models, multi-view geometry, pose estimation, and nonlinear optimization.
• Experience deploying CV/ML pipelines on embedded/edge platforms.
• Experience defining coordinate frames, timestamps, uncertainty and health interfaces for downstream estimation systems.

Responsibilities

• Learned Local Features and Matching.
• EO/IR Data and Model Strategy.
• Geometric Vision and Mathematical Review.
• Navigation-Ready Perception and Integrity.
• Time Synchronization, Calibration, nadir and oblique operation.
• Embedded ML and System Co-Design.
• Technical Leadership and Validation.

Skills

Python

C++

ROS

Deep learning

Computer vision

Linear algebra

Probability & statistics

Optimization

Embedded systems

Dataset design

Education

Bachelor’s degree in Robotics/EE/CS

Master’s degree in Robotics/EE/CS

PhD in Robotics/related field

Tools

PyTorch

OpenCV

ONNX

TensorRT

CUDA

NVIDIA profiling tools

ROS

ROS 2

Docker

Gazebo

V-REP

Unity3D

Job description

• Technical ownership of classical CV and learning-based visual navigation systems for autonomous UAVs operating in GNSS-degraded and denied environments
• Opportunity to define the architecture connecting perception, state estimation, embedded computing and flight systems
• Research and product-development responsibility across EO, IR, inertial, radar and RF-derived information
• Ownership of dataset strategy, mathematical correctness, uncertainty modelling and system-level validation
• Opportunity to lead embedded deployment on resource-constrained computing platforms
• Technical leadership across computer vision, machine learning, navigation, estimation and sensor-integration teams

About the Role:

The Senior Computer Vision & Machine Learning Engineer is a senior hands-on technical role responsible for architecting and delivering learning-based visual navigation and resilient-PNT capabilities.

The engineer will own the technical direction for learned local features, day-night adaptation, geometric vision, vision-aided localization, navigation-ready perception outputs, camera-IMU integration and embedded ML optimization.

The role also includes developing multimodal navigation capabilities and statistical(-learning) methods for quality determination.

The engineer must be able to connect ML performance with geometry, uncertainty, navigation integrity, embedded constraints and field behaviour. This role requires system-level accountability in addition to algorithm development.

Key Responsibilities:

• Learned Local Features and Matching
• EO/IR Data and Model Strategy
• Geometric Vision and Mathematical Review
• Navigation-Ready Perception and Integrity
• Time Synchronization, Calibration, nadir and oblique operation
• Embedded ML and System Co-Design
• Technical Leadership and Validation

Minimum Qualifications:

• Bachelor’s, Master’s or PhD degree in Robotics, Electrical/Electronics Engineering, Computer Science, or a related field
• 6+ years (Bachelor’s) or 5+ years (Master’s) or 1+ years (PhD) or more years of relevant experience; demonstrated architecture ownership and technical depth are more important than a strict year count
• Strong Python, C++, and ROS proficiency
• Advanced practical experience with modern deep-learning frameworks and computer-vision libraries
• Demonstrated experience building visual localization such as visual-inertial odometry, SLAM, scene matching or closely related navigation systems
• Deep understanding of camera models, multi-view geometry, robust pose estimation and nonlinear optimization
• Experience designing datasets and training or adapting models using domain-specific imagery
• Experience defining coordinate frames, timestamps, uncertainty and health interfaces for downstream estimation systems
• Experience deploying CV/ML pipelines on embedded or edge-computing platforms
• Strong applied foundations in linear algebra, probability, statistics, optimization and numerical methods.
• Demonstrated ownership of technical architecture, validation strategy and field-deployed systems
• Ability to mentor engineers and lead cross-functional technical decisions

Mathematical Expectations

Candidates must be capable of applying and reviewing:

• Matrix factorization, eigenvalue problems, SVD and numerical conditioning
• Least squares, weighted least squares, convex and nonlinear optimization
• Rotation matrices, quaternions, SE(3) transformations and Jacobians
• Bayesian estimation, conditional probability and probabilistic graphical reasoning
• Covariance modelling, cross-covariance and uncertainty propagation
• Hypothesis testing, likelihood-ratio testing and confidence calibration
• Sequential methods, change detection and time-series analysis
• Information-based experiment design and observability
• Statistical consistency and false-alarm/detection-probability analysis

Preferred Qualifications:

• Experience with aerial EO, thermal/IR or satellite imagery
• Experience with learned local-feature systems
• Experience with radar-camera-imu calibration or multimodal sensor fusion
• Experience with PyTorch, OpenCV, ONNX, TensorRT, CUDA and NVIDIA profiling tools
• Experience with ROS or ROS 2, Docker and production ML pipelines
• Experience with resource-constrained ARM systems
• Experience with ArduPilot, PX4, MAVLink or autonomous UAV flight stacks
• Experience developing or integrating Kalman filters, factor graphs or nonlinear estimators
• Experience conducting UAV flight tests and defining system-level qualification criteria
• Publications, patents or demonstrated research contributions in visual navigation, multimodal learning or resilient PNT

Additional Considerations for PhD Graduates

Candidates with a PhD may be considered for an enhanced designation or role variant (e.g., Lead Engineer) based on:

• Depth of thesis/research experience in robotics, UAV autonomy, perception, or control systems
• Demonstrated hands-on work in VIO, SLAM, sensor fusion, or advanced multimodal sensor-fusion workflows
• Internships or lab experience involving UAV testing, system integration, and computer vision pipelines
• Ability to take ownership of specific subsystem modules or small projects early in their tenure
• Strong publication track record (IEEE Transactions, ICRA, IROS, CoRL, CVPR, ECCV, NeurIPS, ICML, ICLR)

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

Autonomous Navigation Algorithms. NonLinear Optimization SLAM (Simultaneous Localization and Mapping) algorithms Computer Vision Machine Learning ROS (Robot Operating System) Optimization Technique - Gradient-based methods Robotics simulation software : Gazebo, V-REP, or Unity3D