Systems Software Engineer (Rust, ML Inference)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
Systems Software Engineer (Rust, ML Inference): Developing and optimizing the core real-time inference engine and SDK for Voice AI with an accent on performance, memory efficiency, and cross-platform reliability. Focus on building robust systems-level components, managing ML model execution pipelines, and ensuring low-latency performance in production environments.
Location: Berlin, Germany (Onsite)
Company
is a Voice AI infrastructure startup building the reliability layer for audio, backed by top-tier investors to make voice interfaces measurable and deployable at scale.
What you will do
- Design and optimize systems-level components of the real-time inference engine and SDK.
- Improve performance, memory usage, and stability of model execution pipelines.
- Maintain and expand C API bindings and wrappers for Python, C++, and Node.js.
- Develop DSP modules and infrastructure for streaming audio workloads.
- Build comprehensive testing pipelines, benchmarks, and profiling tools for runtime validation.
- Collaborate with ML researchers to bridge the gap between research models and production-ready deployment.
Requirements
- Strong experience in systems programming with Rust.
- Solid understanding of C/C++ interoperability, ABIs, and FFI design.
- Experience building or maintaining SDKs, inference runtimes, or developer-facing systems.
- Familiarity with real-time systems, performance optimization, and memory management.
- Experience with ML inference runtimes such as ONNX, Burn, or TensorRT.
- Ability to work onsite in Berlin.
Culture & Benefits
- Competitive salary package with equity and stock options.
- Dynamic, fast-paced startup environment with high ownership and autonomy.
- Opportunity to work with a world-class team on groundbreaking Voice AI technology.
- Direct impact on the future of audio interfaces and real-world production systems.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →