Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$51к$69кмедиана$99к
21 в выборке
На СНГ рынке: $3.6к/мес ($3к - $4.7к)
88
Крутая вакансия
развернуть
Роль сочетает исследования в области речи и мультимодального ИИ с разработкой надежных инженерных систем. Необычная тема эмоционального благополучия интересна, но о компании и долгосрочном развитии продукта пока мало информации.
Передовой ИИ
Четкая зона ответственности
Современный стек
Гибкий подход
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Research Engineer (AI) (speech, audio and multimodal machine learning): Building and improving models for SOUL, Nightlet’s State Observation & Understanding Layer, with an accent on reproducible training, inference, benchmarking and large-scale data workflows. Focus on designing experiments, evaluating model quality and robustness, and turning open-ended research into validated model improvements.
Location: Remote within the EU; Hamburg office optional.
Company
hirify.global builds emotional intelligence products combining speech, audio, multimodal signals, body data and astronomical context through an app and optional bracelet.
What you will do
Investigate speech-language, audio and multimodal machine learning questions and design experiments.
Train, fine-tune, validate and improve in-house and partner models.
Build reliable software for distributed training, inference and reproducible benchmarking.
Develop workflows for collecting, storing, preparing and analyzing large datasets.
Define evaluation protocols covering quality, generalization, robustness and research progress.
Analyze datasets, model outputs and failure cases, collaborate with partner research teams, and document experiments, results, limitations and technical decisions.
Requirements
At least two years contributing to model training or fine-tuning on large text, audio, image or video datasets, with substantial experience in speech, audio or multimodal machine learning.
Strong Python and PyTorch skills, including developing, debugging and maintaining training and evaluation code.
Experience writing robust, maintainable research-system code beyond notebooks and prototypes.
Evidence of research and model-building ability through publications, model releases, shipped systems or substantial open-source work.
Ability to iterate on uncertain research directions, own open-ended technical problems, and communicate methods, findings and limitations clearly to research and partner engineering teams.
A PhD is not required; demonstrated research ability and the quality of contributed models, experiments and systems matter.
Nice to have
Experience training or fine-tuning transformer models.
Depth in text-to-speech, speech recognition, speech-to-speech, prosody, emotion recognition or audio understanding.
Experience with post-training, preference learning or human evaluation of generative models.
Experience improving the efficiency of distributed training, inference or large-scale data processing.
Collaboration with external research teams on model development or evaluation.
Culture & Benefits
Flexible work culture without hour tracking.
Unlimited sport benefit and a spiritual allowance covering activities such as meditation or therapy.
Models are used by the team regularly in addition to benchmark evaluation.
Collaborative culture focused on honest feedback and testing the products internally.
Hiring process
Submit a CV, a link to something you have built and a few honest sentences through the application.
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 →