Senior Perception Engineer-AI/ML

Overview

Develop and deploy robust perception solutions with sensor fusion for autonomous vehicles.

Key Responsibilities

• sensor fusion
• perception testing
• data filtering
• calibration workflow
• c++ development
• system integration

Tasks

-Implement, validate, integrate and deploy robust perception solutions with sensor fusion for autonomous vehicles using LiDAR, Camera, Radar and GNSS sensors. -Test, validate, and fine-tune perception modules independently to handle edge cases, unusual driving conditions, and rare corner scenarios. -Build pre-processing and post-processing filters to clean and refine raw sensor data for downstream perception modules. -Develop and maintain intrinsic and extrinsic sensor calibration workflows -Develop robust, production-ready C++ and Python code and integrate it into production pipelines -Independently design experiments, debug challenging issues, and verify solutions in simulation, test bench and on-vehicle tests. -Collaborate with cross-functional teams to ensure smooth integration of perception, calibration, and localization functions with planning and control sub-systems. -Elicitation of system requirements to derive ODD and KPI goals and instructions to capture datasets for training, validation and testing.

Requirements

• ros2
• docker compose
• c++
• python
• cuda
• lidar

What You Bring

-Knowledgeable in ROS2 software stack and Docker Compose tool for integration and deployment of perception algorithms. -Hands on knowledge in sensor fusion techniques (Kalman Filters, Extended Kalman Filters, Unscented Kalman Filters, Particle Filters) for combining data from LiDAR, camera, radar, and GNSS. -Knowledge of parallel computing (CUDA, OpenCL, OpenMP) for real-time acceleration. -Solid understanding of LiDAR, camera, RADAR, and GNSS sensor data, their characteristics and calibration mechanisms (intrinsic and extrinsic) -Excellent debugging and problem-solving skills. -Strong hands-on programming skills in C++ and Python. -Optimize perception algorithms for real-world deployment, balancing accuracy, latency, and computational performance on embedded systems. -Prior experience with automotive-grade software development practices. -Familiarity with SLAM, visual-inertial localization, or GNSS-INS integration. -Proven experience developing perception algorithms for autonomous vehicles or robotics covering object detection, semantic segmentation, tracking and localization using classic methods (e.g. RANSAC, Euclidean clustering, Occupancy Grid Mapping) and deep learning models (e.g. PointNet++, PointPillars) for Lidar point clouds and radar scans. -Experience in implementing computational principles effectively using libraries like Eigen, Boost, GeographicLib etc. -Good understanding of real-time system constraints and optimization trade-offs for product-level deployment. -Ability to think independently, handle unusual edge conditions, and fine-tune algorithms for real-world robustness. -Familiarity with object detection models (e.g., YOLO, Faster R-CNN, SSD) and point cloud libraries (PCL).

People Also Searched For

About Quest Global

-By blending deep technical know‑how with digital innovation, it evolved into a global engineering partner. -Projects range from next‑gen automotive safety systems and AI‑driven semiconductors to cutting‑edge aerospace and defense platforms. -Its financial growth is steady—built on recurring partnerships with Fortune‑top‑10 clients across key sectors. -A hallmark project: sustaining global EMS/ADMS energy‑management software for a major utilities provider. -Another standout: accelerating MRI development and support globally for a leading MedTech firm. -Its edge lies in full‑lifecycle delivery—concept, design, embedded software, manufacturing support, digital sustenance. -Known for foresight, the firm not only solves challenges but anticipates and prevents them.

Sector Specialisms