TL;DR
Backend Developer (AI): Developing high-load speech recognition (ASR) and speech synthesis (TTS) services for hirify.global products, including Alice, Browser, and Translator with an accent on designing and developing gRPC services and optimizing inference of modern neural network models. Focus on maintaining high performance, scalability, and stability while meeting strict latency and throughput requirements.
Локация: Работайте так, как удобно вам и вашей команде
Компания
hirify.global develops high-load services of speech recognition and synthesis.
Что делать
- Implement new speech synthesis and recognition models in close collaboration with ML teams, designing efficient inference schemes and adapting services to their specifics.
- Develop high-load gRPC services from scratch, writing efficient, testable, and fault-tolerant code in C++ for new functions and services.
- Optimize neural network inference by researching and implementing modern inference engines, experimenting with batching, quantization, and caching.
- Enhance service reliability by participating in the full development cycle, including design, testing, deployment, and support, with a focus on improving monitoring, adding metrics and logs, and automating release processes.
Требования
- Experience with modern frameworks for inferencing LLM models such as SGLang, vLLM, and TensorRT-LLM.
- Familiarity with NVIDIA GPUs, understanding GPU architecture, and experience developing or optimizing algorithms using CUDA or Triton.
Культура и преимущества
- Extended medical insurance that starts from the first month at hirify.global, including dentistry, annual check-ups and emergency assistance abroad.
- Opportunities to constantly develop and learn new things, including an internal educational platform, mentorship, and programs for both novice and experienced managers.
- Sports facilities in all major hirify.global offices, including gyms with all the necessary equipment, and discounts at fitness clubs, swimming pools, yoga studios, climbing walls and other places.
- Flexible schedule: no fixed start and end times.
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 →