3D Simulation & Data Pipeline Engineer

Stellenbeschreibung:

Overview Get AI-powered advice on this job and more exclusive features. Design the digital environments that power autonomous drone systems! About SkyBound and the job: SkyBound Rescuer is dedicated to automating drones for rapid response, optimised performance, and better cost-effectiveness, ultimately saving, protecting, and improving lives. We're seeking a 3D Simulation & Data Pipeline Engineer to help turn aerial imagery into interactive, high-fidelity simulation environments - used for autonomy testing, training, and synthetic data generation. In this hands-on role, you'll collaborate closely with the Data & AI team, owning the pipeline from photogrammetry to Unreal Engine 5, and supporting workflows for sensor emulation, automated annotation, and scenario design. Our team is lean, technically strong, and motivated by impact. We operate remotely, with monthly on-site collaboration in Southampton. Work Location: Hybrid - Remote / Southampton SO19 Reports to: Head of Data and AI

Simulation Environment Design

Build and maintain high-fidelity 3D scenes in Unreal Engine 5, incorporating realistic lighting, weather systems, and sensor placement.
Create and configure simulation actors and agents to replicate target behaviors or real-world phenomena (e.g. people, vehicles, crowds) using Unreal's Actor framework.
Implement automation workflows for scene management, data capture, and sensor simulation using Blueprints and Python APIs.
Procedural generation of environments to augment datasets (people and landscapes).

Photogrammetry and Scene Reconstruction

Develop pipelines to ingest and transform aerial drone imagery and photogrammetric reconstructions into usable 3D assets.
Polish and optimise large-scale mesh models using tools such as Blender, reducing blockiness and improving terrain/texturing fidelity.
Apply GIS and georeferencing knowledge to align models accurately within simulation space.

Synthetic Data & ML Pipeline Support

Enable capture and export of synthetic data (e.g., RGB, depth, segmentation, metadata) for use in machine learning pipelines.
Collaborate directly with ML researchers to tailor simulation scenarios for self-supervised, multi-frame, and temporal learning strategies.

SITL/HITL Integration

Extend existing Software-in-the-Loop (SITL) and Hardware-in-the-Loop (HITL) frameworks to support testing of perception and planning systems.
Contribute to sensor emulation, camera (RGB and Thermal/IR), and realtime control integration.

Required Skills and Experience

3-5 years of hands-on experience in simulation engineering, 3D environment construction, or synthetic data workflows - strong candidates with less experience but strong portfolios will also be considered.
Strong experience with Unreal Engine 5, including Blueprint scripting and level/actor design.
Proficiency in Python, especially for simulation automation and data pipeline tasks.
Hands-on experience with Blender or similar tools for mesh editing and optimisation.
Familiarity with reality capture workflows, including drone-based data acquisition and tools such as RealityCapture.
Experience with photogrammetry, drone autonomy, or 3D modelling from aerial imagery.
Basic understanding of SITL/HITL architectures and how they interface with tools like PX4, ROS, or MAVLink
Familiarity with real-time rendering and performance optimisation for simulation environments.
Experience with procedural scene generation or automated 3D asset pipelines.
Basic familiarity with MATLAB and Simulink, especially in the context of robotics, control systems, or simulation validation.
Working knowledge of GIS systems, including geospatial data formats (GeoTIFF, DEM, shapefiles) and coordinate transformations.
Experience with synthetic data generation for ML, including sensor simulation and annotation export.
Fluency in English is essential.
Flexible and a team player.