AI Developer (Signal Processing & Speech Technology)
CareerFinders Recruitment Services
Location
Limassol
Hybrid
Type
full-time
Salary (Gross)
Salary upon interview
Apply within
17 days
Category
AI & Machine Learning
About the Role
CareerFinders is assisting an international developer of sophisticated software solutions, with regional operations in Cyprus, in recruiting an AI Developer specializing in Signal Processing & Speech Technology for their growing team in Limassol. The successful applicant will be primarily responsible for the selection, adaptation, and fine-tuning of models for signal and speech processing. The ideal candidate will have an academic background in Computer Science, Electrical Engineering, Physics, Mathematics, or a related field, coupled with multiple years of experience using PyTorch. Solid C++ knowledge is essential, along with experience with the PyTorch C++ API. Experience with CUDA kernel development or custom Triton backends is a plus. Excellent verbal and written English communication skills are required. The client offers an attractive remuneration and benefits package based on skills and experience.
Responsibilities
- Selection, adaptation, and fine-tuning of models for signal and speech processing (including CNN/Transformer architectures, state-space models, and end-to-end ASR)
- Optimizing inference: quantization, pruning, kernel fusion, batching strategies, mixed precision, TorchScript/torch.compile, export to ONNX/TensorRT
- Implementing high-performance inference paths in PyTorch and LibTorch (PyTorch C++)
- Deployment and operation of models in the NVIDIA Triton Inference Server—including model repository, ensembles, business logic scripting, dynamic batching, and custom backends
- Building processing chains around the models: signal preprocessing, feature extraction, streaming pipelines, pre-/post-processing
- Taking prototypes from development to market readiness: benchmarking, profiling, test coverage, reproducibility, monitoring, and versioning of models
- Close collaboration with system architecture and product management
- Evaluating new methods and tools for their practical suitability for performance goals
Requirements
- BSc/MSc in Computer Science, Electrical Engineering, Physics, Mathematics, or comparable qualification/equivalent practical experience
- Several years of experience with PyTorch beyond training, understanding model slowdowns in production and how to fix them
- Solid knowledge of C++ (C++17 or later) and experience with the PyTorch C++ API
- Practical experience with Triton Inference Server
- In-depth knowledge of digital signal processing (FFT, filter design, modulation, spectrogram representations) or speech processing (ASR, speaker embeddings, speech enhancement)
- Experience with profiling and GPU optimization (CUDA fundamentals)
- Proficiency with Linux, Git, Docker, and CI/CD
- Fluent in spoken and written English
- Experience with CUDA kernel development or custom Triton backends (Nice to Have)
- Experience working with SDR hardware and real-world RF datasets (Nice to Have)
Benefits
- ✓Attractive remuneration package
- ✓Private health insurance
- ✓Provident fund
- ✓Gym membership
- ✓Hybrid working scheme
- ✓Continuous personal development scheme
Know someone who'd be a great fit?
Share this role with a friend, ex-colleague or your network. They'll thank you.