Software Engineer, AI Frameworks, Mid
Guarda esta oferta y sigue tu búsqueda
Crea una cuenta gratis para guardar empleos, crear alertas y volver a esta oferta desde tu panel.
Al continuar, aceptas nuestros Términos & Política de Privacidad.
Job Description
Our partner is looking for Mid and Senior Software Engineers to develop software components, AI frameworks and execution environments that integrate closely with advanced processor hardware.
The role focuses on modern C++, AI model execution, software performance and hardware acceleration, working across low level system interfaces and higher level AI frameworks. You will collaborate closely with hardware, compiler and AI engineering teams to optimize inference workloads, improve software performance and support the integration of AI models with custom processor architectures.
Depending on your experience, you may contribute to AI framework integration, model optimization, compiler infrastructure, kernel development or performance tooling.
Location: Barcelona, Spain, Hybrid
Level: Mid / Senior
Responsibilities
Design and develop software components using modern C++ and Python
Develop and integrate AI frameworks and execution runtimes with advanced processor hardware
Work across low level system interfaces, compiler infrastructure and higher level AI frameworks
Support the execution and optimization of AI models on specialized hardware accelerators
Profile, debug and optimize complex software and AI inference workloads
Analyze model execution to identify performance bottlenecks related to computation, memory bandwidth, data movement and hardware utilization
Improve software performance, scalability, reliability and execution efficiency
Support integration between AI frameworks, optimized kernels and processor hardware
Contribute to software architecture, technical design and performance optimization strategies
Collaborate closely with hardware, compiler, kernel development and AI engineering teams
Investigate model execution issues, compatibility problems and performance limitations
Contribute to the development of profiling tools, debugging infrastructure and software optimization techniques
Requirements
Strong professional experience with modern C++
Good knowledge of C
Working knowledge of Python
Strong experience working in Linux environments
Experience developing multithreaded software, including concurrency and synchronization
Hands on experience with software profiling and debugging, including GDB or similar tools
Experience working on complex software systems with multiple interacting components
Strong understanding of software performance, memory management and optimization
Experience analyzing performance bottlenecks and identifying their underlying causes
Ability to explain technical optimization decisions and their impact on software performance
Familiarity with how AI frameworks, software runtimes or computational workloads interact with hardware
Strong problem solving skills and attention to detail
English level C1
Optional / Desired
RISC V experience
Understanding of AI model inference, including transformer architectures, attention mechanisms and model execution
Knowledge of LLM inference concepts, including KV cache, prefill, decode and PagedAttention
Experience with AI model optimization techniques, including quantization, pruning, operator fusion and model compression
Understanding of static and dynamic quantization, precision formats and their impact on model performance
Experience profiling AI workloads to identify compute, memory bandwidth and data transfer bottlenecks
PyTorch dispatcher or ATen internals
vLLM or SGLang
ONNX Runtime or OpenVINO
TensorFlow or other AI execution frameworks
CMake
LLVM or GCC
MLIR, intermediate representations or compiler graph lowering
Hardware and software interfacing
Multi GPU deployment
C/C++ and Python interoperability, including pybind11
Open source contributions
CUDA or ROCm
AI kernel development and optimization
Experience integrating mathematical operations and kernels into AI execution frameworks
Hardware emulation, NPU or GPU profiling
Bare metal programming
AI application development
Workload analysis and performance tuning
Experience with custom processors, NPUs, GPUs or other hardware accelerators
What’s in it for you?
Flexible working hours
Hybrid model, 4 days office, 1 day from home
1 week per year working from anywhere
25 days annual leave plus December 24 and 31
Monthly food allowance
Private medical insurance
Relocation bonus
Housing search support
Visa support if required
Family relocation support
Virtual shares
Language classes
Tax incentive, subject to eligibility