Software Engineer
Pubblicato il 15-08-2026 - Altro in Leivi
Careers Software Engineer (Mid-Level) – AI Systems
Engineering Hybrid / Europe / Italy Full-time
Location: Remote (EU preferred) Experience: 2–5 years
About Us Daidalos is a deep-tech company developing a proprietary AI accelerator IP designed for efficient and deterministic AI execution.
We are working at the intersection of AI, hardware design, and embedded systems, building architectures that go beyond traditional GPU-based approaches.
The role This role goes beyond traditional software engineering.
You will work on mapping AI models onto custom hardware, optimizing execution, and contributing directly to hardware-software co-design.
We are not looking for someone who simply uses frameworks — we are looking for someone who wants to understand how they work under the hood and improve them.
What you will work on Optimization and deployment of AI models (PyTorch / ONNX) on custom architectures
Development in C++ / C / Python with strong focus on performance
Hardware-software co-design with FPGA-based systems
Optimization under real-world constraints (memory, latency, throughput)
Contribution to runtime and execution pipelines
Requirements 2–5 years of experience in software engineering
Strong skills in C++ and/or Python
Experience with performance-critical code
Understanding of system-level behavior (memory, compute, latency)
Nice to have Experience with AI frameworks (PyTorch, ONNX, TensorRT, etc.)
Exposure to embedded systems or hardware acceleration
Experience with profiling and optimization tools
Familiarity with Linux and low-level programming
What we are looking for Strong problem-solving skills
Ability to work across abstraction layers
Interest in hardware-aware software design
Ownership and proactive mindset
What we offer Full remote work
High-impact role in a deep-tech startup
Direct involvement in architecture-level decisions
Opportunity to work at the boundary of AI and hardware
Equal Opportunity Daidalos is committed to equal employment opportunities and