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
$218к$250кмедиана$294к
50 в выборке
На СНГ рынке: $5.1к/мес ($3.8к - $7.4к)
85
Хорошая вакансия
развернуть
Роль четко определена с ясным акцентом на мультимодальный ИИ, а компания находится в трендовом домене с хорошей поддержкой, но отсутствие информации о зарплате является недостатком.
Трендовый домен
Четко определенная роль
Нет информации о зарплате
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Technical Lead, Multimodal Research (AI): Building multimodal video-understanding infrastructure that indexes physical-AI datasets across video, lidar, radar, and simulation data with an accent on model selection, evaluation, and production inference at petabyte scale. Focus on designing cost-efficient model cascades and routing, defining evaluation standards, and translating customer research needs into scalable datasets and technical programs.
Location: San Francisco, United States; 4 days/week in the SF Mission District office
Company
hirify.global builds open-source and commercial infrastructure for multimodal AI, enabling Physical AI companies to search large-scale video and sensor corpora and create training datasets or actionable alerts.
What you will do
Own the modeling strategy across the platform, including model families, representations, training approaches, and research priorities.
Move computer vision and multimodal approaches from prototypes to production inference across hundreds of thousands of video hours and petabyte-scale corpora.
Define model benchmarks, evaluation standards, and customer acceptance criteria.
Optimize the cost and performance of video understanding through distillation, cascades, routing, quantization, GPU utilization, and throughput improvements.
Translate customer research needs into technical programs covering taxonomies, model plans, datasets, and quality instrumentation.
Set the technical direction for multimodal research as a senior individual contributor without direct reports.
Requirements
5+ years of experience in applied computer vision or multimodal machine learning.
PhD or MS in computer science, electrical engineering, robotics, or applied mathematics with a computer vision or machine learning focus, or a comparable publication or production record.
Strong knowledge of VLMs, VQA, embeddings, representation learning, detection, tracking, segmentation, and retrieval.
Hands-on experience training and evaluating models at scale on real video and sensor data, including PyTorch prototyping, inference performance, GPU utilization, throughput, and cost.
Experience in perception or multimodal work at a self-driving, robotics, Physical AI, frontier research, or visual-data company.
Nice to have
Publications at CVPR, ICCV, ECCV, NeurIPS, ICML, or ICLR.
Experience building or fine-tuning VLMs or multimodal foundation models.
Experience with long-form video, temporal reasoning, embeddings, retrieval, or content-aware indexing at scale.
Experience with lidar, radar, depth, simulation output, evaluation frameworks, labeling taxonomies, annotation programs, or large-scale GPU optimization.
Culture & Benefits
In-person, tight-knit team working 4 days per week from the San Francisco office.
Competitive compensation and startup equity.
Health, vision, dental, and 401(k) coverage with company match.
Flexible PTO, commuter benefit, and catered lunches and dinners for San Francisco employees.
Latest Apple equipment, team-building events, and poker nights.
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 →