EdgeAI Engineer

Company Description

Aistech Space is focused on generating affordable, recurrent, high-resolution thermal imagery of the planet to provide a new perspective of Earth’s changing resources. Based in Barcelona, we are revolutionizing remote sensing through advanced space technologies and artificial intelligence, delivering innovative solutions for environmental monitoring and resource management.

Aistech Space is seeking a highly specialized Edge AI Engineer to serve as the critical bridge between Data Science, Software, and Hardware Engineering teams. This role is focused on the successful integration, deployment, and performance optimization of machine learning models across highly constrained operational environments, including in-orbit systems (satellites, Coral TPUs, AMD Versal SoCs), ground processing infrastructure, and cloud-based platforms.

You will drive the complete deployment lifecycle of AI models, ensuring our space-borne and ground-based AI systems are reliable, ultra-efficient, and optimized for real-time execution. Working closely with multidisciplinary engineering teams, you will transform research models into operational flight software while continuously evaluating new hardware accelerators and Edge AI technologies to expand Aistech Space’s capabilities.

Education: Master’s or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related technical field.

Languages: Fluency in English.

Embedded

Experience:

2+ years of professional experience in Embedded Software Development or Hardware-Software Co-design.

Programming: High proficiency in C/C++ for low-latency, real-time embedded execution, and Python for scripting, model conversion, and prototyping.

Development Environment: Strong experience with Linux environments, cross-compilation tool chains, and collaborative development using Git/Git Hub.

Hardware Acceleration: Hands-on experience working with Neural Processing Units (NPUs / Edge TPUs) and Field Programmable Gate Arrays (FPGAs / Adaptive SoCs).

Deployment Frameworks: Strong expertise in ML deployment and optimization frameworks such as Tensor Flow Lite (including the Edge TPU Compiler), ONNX Runtime, or Vitis AI.

Model Optimization: Solid understanding of INT8 quantization, pruning, and weight compression techniques for edge deployment.

AMD Versal & Vitis: Experience with AMD Versal AI Engines, Vitis AI, Vitis Model Composer, Vivado, DPU integration, AIE-ML kernel optimization, and hardware-level performance analysis.

Hybrid Cloud/Edge MLOps: Experience deploying AI models through APIs and web services (FastAPI, Flask, gRPC) combined with cloud infrastructure and containerization technologies such as Docker, Kubernetes, AWS, or GCP.

High-Performance Computing: Familiarity with HPC environments and workload management systems such as SLURM.

• Familiarity with lightweight containerization technologies for constrained systems (e.g., Singularity, microcontainers).
• Knowledge of advanced data compression algorithms and telemetry constraints for in-orbit data transmission.
• Previous experience within the aerospace, New Space, or remote sensing industries.

• Collaborate closely with FPGA, Embedded Software, and Data Science teams to deploy, profile, and maintain AI solutions on satellites and other constrained Edge AI platforms such as Google Coral and AMD Versal.
• Convert and optimize machine learning algorithms developed in Python into high-performance, deterministic C/C++ implementations suitable for embedded execution.
• Design, build, and maintain end-to-end ML deployment pipelines spanning cloud-based training environments through in-orbit edge execution.
• Resear…