Collaborate with research peers to conceive and define novel initiatives in large language model development
Conduct comprehensive reviews of academic and technical literature to assess, categorize, and synthesize current methodologies
Develop and execute experimental frameworks tailored to large language models
Lead transformer-based model training workflows on distributed computing systems and oversee deployment of new models into production environments
Communicate technical findings and complex concepts clearly and effectively to relevant stakeholders
Maintain continuous awareness of advancements in deep learning and LLM technologies, with focus on practical integration into product offerings

Comprehensive medical, dental, and vision coverage
Annual allowance for wellness-related expenses
Access to mental health resources and support services
Life, short-term disability, long-term disability, and income protection insurance
Unlimited paid time off
Extensive paid leave for new parents
Flexible working hours to support individual productivity
Twelve paid U.S. public holidays annually
Quarterly stipend for personal productivity tools or services
One-time financial support for home office setup
401(k) retirement plan with employer matching contributions
Eligibility for tax-advantaged savings programs
Stipend for professional development and educational pursuits
Opportunities to attend industry talks and conferences
Support for participation in employee resource groups
Regular AI enablement workshops and training sessions
Region-specific benefits administration for international employees via Employer of Record model

Not specified

Flexible schedule; remote options available

AI-first research team focused on advancing voice-native foundation models

The organization leads in powering the growing Voice AI economy, offering real-time APIs for speech-to-text, text-to-speech, and scalable voice agent development
Over 200,000 developers and more than 1,300 organizations leverage its technology, including notable companies such as Twilio, Cloudflare, and Sierra
Its foundation models for voice are accessible via cloud APIs or self-hosted deployments, delivering high accuracy, low latency, and cost efficiency
Backed by a Series C round from top-tier global investors, the platform has processed over 50,000 years of audio and transcribed more than 1 trillion words
No other entity possesses deeper expertise in voice technology

An AI-first mindset is fundamental—using and advancing AI is central to operations, innovation, and performance evaluation
All team members are expected to actively integrate and experiment with advanced AI tools in daily workflows
Success is measured by effective application of AI, with emphasis on consistent and inventive use of emerging capabilities
Candidates must adapt quickly to new models and incorporate AI into evolving processes
The pace of work mirrors the speed of AI advancement, requiring constant evolution and agility
This role is unsuitable for those seeking rigid, traditional work structures or who are hesitant to experiment and learn continuously

Voice remains the most intuitive interface for human-machine interaction
Current sequence modeling approaches, reliant on scaling models and data together, fall short of enabling universal voice AI
Core challenges stem from audio data scarcity and extreme diversity across voices, speaking styles, and acoustic environments
Even with vast audio datasets, high dimensionality leads to prohibitive computational and storage demands at global scale
New paradigms in audio AI are essential to overcome these barriers and make voice interaction universally accessible

We seek researchers who view unsolved problems as opportunities to pioneer novel solutions
Ability to design and run decisive experiments that validate or disprove ideas within days
Vision to scale proven concepts by orders of magnitude
Strong drive to use AI for automating and amplifying personal impact
Ideal candidates feel energized by these expectations and generate multiple ideas upon reading the role description
The position requires deep problem obsession, creative methodology, and relentless pursuit of elegant, scalable outcomes
Technical hurdles are significant, but the potential for transformative impact is substantial

Are passionate about artificial intelligence and excited to work on cutting-edge LLM research
Are interested in generating and applying new scientific insights to develop and deploy large language models
Enjoy building systems from scratch and creating novel technical architectures
Possess strong communication skills and can clearly explain complex technical concepts
Are highly analytical and thrive when conducting in-depth technical analysis

Supported by leading investors such as Y Combinator, Madrona, Tiger Global, Wing VC, and NVIDIA, with over $215 million in total funding raised

The company is committed to equal opportunity employment, welcoming diverse voices and perspectives, and fostering a collaborative, customer-first culture that values growth and speed; it does not discriminate based on race, religion, color, national origin, gender, sexual orientation, gender identity, age, marital status, veteran status, disability, pregnancy, parental status, genetic information, political affiliation, or other protected statuses
Accommodations are available for applicants who request them

For candidates outside the U.S., an Employer of Record model is used in many countries, with benefits administered locally under regional regulations
Benefits vary by region—some international employees may receive benefits not available to U.S. staff, and vice versa
As the organization grows, it will continue assessing opportunities for greater benefits alignment, though identical global structures are not always legally or operationally feasible

Not specified

Deepgram is hiring a Research Staff, LLMs