- Страна
- США
Откликайтесь
на вакансии с ИИ
Member of Technical Staff - Post Training, Applied (Audio)
Оценка ИИИсключительная возможность работать в стартапе-единороге, вышедшем из MIT, над передовыми Liquid Foundation Models. Высокая автономность, отличный соцпакет и работа на стыке науки и реального внедрения.
Сложность вакансии
Оценка ИИРоль требует глубокой экспертизы в узкой нише аудио-ML (ASR, TTS) и умения работать с ограничениями реальных устройств. Кандидат должен сочетать навыки исследователя и инженера, самостоятельно ведя проекты от данных до деплоя.
Анализ зарплаты
Оценка ИИДля позиции уровня MTS в AI-стартапе стадии 'unicorn' в Сан-Франциско рыночные зарплаты обычно начинаются от $180k. Учитывая дефицитность специалистов по аудио-ML, итоговое предложение может быть значительно выше за счет опционов.
Сопроводительное письмо
I am writing to express my strong interest in the Member of Technical Staff - Post Training (Audio) position at Liquid AI. With a deep background in speech processing and machine learning, I am particularly drawn to Liquid AI's mission of building efficient, general-purpose AI systems that excel under real-world hardware constraints. My experience in fine-tuning ASR and TTS models, combined with a rigorous approach to data quality and evaluation, aligns perfectly with your goal of delivering high-performance audio solutions to enterprise customers.
In my previous work, I have managed end-to-end ML pipelines, from designing custom data augmentation strategies for noisy environments to implementing SFT and preference alignment techniques. I thrive in environments where low latency and memory efficiency are treated as primary design goals rather than afterthoughts. I am excited by the prospect of acting as a technical bridge for your customers and building the reusable tooling that will scale Liquid AI’s applied audio capabilities.
Составьте идеальное письмо к вакансии с ИИ-агентом
Откликнитесь в liquid-ai уже сейчас
Присоединяйтесь к Liquid AI, чтобы создавать передовые аудио-модели для реальных устройств и определять будущее ИИ вне облака!
Описание вакансии
About Liquid AI
Spun out of MIT CSAIL, we build general-purpose AI systems that run efficiently across deployment targets, from data center accelerators to on-device hardware, ensuring low latency, minimal memory usage, privacy, and reliability. We partner with enterprises across consumer electronics, automotive, life sciences, and financial services. We are scaling rapidly and need exceptional people to help us get there.
The Opportunity
This is a rare chance to own applied post-training work end-to-end for audio workloads, adapting Liquid Foundation Models for customers who need speech and audio capabilities that run on-device under real-time constraints.
You will act as the technical bridge between customer requirements and model delivery for audio tasks. You will lead engagements from scoping through evaluation, with full ownership over how audio models are adapted and shipped. Between engagements, you will build reusable applied workflows and tooling that accelerate future delivery.
If you care about audio data quality, speech model evaluation, and making audio models actually work in production for real customers, this is the role.
What We’re Looking For
We need someone who:
- Takes ownership: Owns customer post-training projects end-to-end for audio workloads, from requirements through delivery and evaluation.
- Thinks end-to-end: Can reason across audio data pipelines, speech-text alignment, model adaptation, and evaluation as a connected system.
- Is pragmatic: Optimizes for model quality and customer outcomes over publications or theory.
- Thrives under constraints: On-device, low-latency, memory-limited audio systems excite you. You see constraints as design parameters, not blockers.
The Work
- Act as the technical owner for enterprise customer post-training engagements involving audio and speech workloads
- Translate customer requirements into concrete post-training specifications for ASR, TTS, and speech-to-speech tasks
- Design and execute data generation, preprocessing, augmentation, and quality filtering processes for audio corpora
- Fine-tune and adapt audio/speech models for customer-specific use cases, owning delivery from requirements through deployment
- Design task-specific evaluations for audio model performance (noise robustness, speaker variation, latency) and interpret results
- Build reusable applied tooling and workflows that accelerate future customer engagements
Desired Experience
Must-have:
- Hands-on experience with data generation and evaluation for ML model post-training
- Experience training or fine-tuning models using SFT, preference alignment, and/or RL
- Strong intuition for data quality and evaluation design
- Experience with speech or audio ML models (ASR, TTS, audio understanding, vocoders, or speech-to-speech systems)
- Proficiency in Python and PyTorch with autonomous coding and debugging ability
Nice-to-have:
- Experience with audio data pipelines at scale (preprocessing, augmentation, quality filtering)
- Experience delivering applied ML work to external customers with measurable outcomes
- Familiarity with on-device deployment under latency and memory constraints
What Success Looks Like (Year One)
- Independently owns and delivers enterprise post-training projects for audio workloads with minimal oversight
- Is trusted by customers as the technical owner for audio engagements, demonstrating strong judgment and delivery quality
- Has built reusable applied workflows or tooling that accelerate future customer engagements
What We Offer
- Real ML work: You will fine-tune audio and speech models, build audio data pipelines, and ship solutions to enterprise customers under real-time on-device constraints.
- Compensation: Competitive base salary with equity in a unicorn-stage company
- Health: We pay 100% of medical, dental, and vision premiums for employees and dependents
- Financial: 401(k) matching up to 4% of base pay
- Time Off: Unlimited PTO plus company-wide Refill Days throughout the year
Создайте идеальное резюме с помощью ИИ-агента
Навыки
- Python
- PyTorch
- Machine Learning
- Deep Learning
- Signal Processing
- Supervised Fine-Tuning
- Reinforcement Learning from Human Feedback
- Data Augmentation
- Text-to-Speech
- Automatic Speech Recognition
Возможные вопросы на собеседовании
Проверка понимания специфики работы с аудио-данными для обучения моделей.
Как бы вы спроектировали пайплайн фильтрации качества для многоязычного аудио-корпуса объемом 10 000 часов?
Важно для работы с клиентами Liquid AI, где важна работа на устройствах.
Какие техники оптимизации (квантование, дистилляция) вы применяли для запуска аудио-моделей в реальном времени на мобильных чипах?
Оценка опыта в Post-Training, указанного в требованиях.
В чем заключаются основные сложности при использовании RLHF или DPO для улучшения естественности речи в TTS-системах?
Проверка навыков оценки качества, критически важных для роли.
Как объективно измерить робастность ASR-модели к фоновому шуму и разным акцентам без использования только WER?
Проверка умения работать с клиентами и переводить бизнес-задачи в технические.
Опишите случай, когда требования клиента противоречили техническим ограничениям устройства. Как вы адаптировали архитектуру модели?
Похожие вакансии
MLOps Engineer (Python)
AI Engineer (CV & Navigation)
Middle, Middle+, Senior GenAI/LLM Разработчик
Middle / Senior GenAI Engineer (CV)
AI Engineer / AI Mentor
Junior разработчик agent AI-систем
1000+ офферов получено
Устали искать работу? Мы найдём её за вас
Quick Offer улучшит ваше резюме, подберёт лучшие вакансии и откликнется за вас. Результат — в 3 раза больше приглашений на собеседования и никакой рутины!
- Страна
- США