--- title: 'Research Staff, LLMs at Deepgram' canonical: 'https://feeny.ai/job/research-staff-llms-deepgram-united-states-k68rft8xvc4q' type: 'job' last_seen: '2026-09-06' --- # Research Staff, LLMs at Deepgram - **Company:** Deepgram - **Location:** United States - **Compensation:** $150k–$250k - **Employment:** full-time - **Work type:** remote - **Posted:** 2026-08-23 - **Last confirmed live:** 2026-09-06 - **Apply:** https://jobs.ashbyhq.com/deepgram/39c2b79b-0269-4711-9354-be5ccf747a98/application **Skills:** Large Language Models (LLMs), Transformer Architecture, Python, PyTorch, Distributed Training, Deep Learning, Data Curation, Experimental Design, Causal LMs, Distributed Inference Schemes, RLHF Labeling, Transformers, Auto-regressive Models, Sequence-to-sequence Models > Deepgram seeks an experienced Research Staff member to join the Research Team, focusing on Large Language Models (LLMs). The role involves defining new research initiatives, conducting experimental programs, optimizing transformer architectures, and deploying models on distributed infrastructure. Candidates must hav... ## Job description ## COMPANY OVERVIEW Deepgram is the leading platform underpinning the emerging trillion-dollar Voice AI economy, providing real-time APIs for speech-to-text (STT), text-to-speech (TTS), and building production-grade voice agents at scale. More than 200,000 developers and 1,300+ organizations build voice offerings that are ‘Powered by Deepgram’, including Twilio, Cloudflare, Sierra, Decagon, Vapi, Daily, Cresta, Granola, and Jack in the Box. Deepgram’s voice-native foundation models are accessed through cloud APIs or as self-hosted and on-premises software, with unmatched accuracy, low latency, and cost efficiency. Backed by a recent Series C led by leading global investors and strategic partners, Deepgram has processed over 50,000 years of audio and transcribed more than 1 trillion words. There is no organization in the world that understands voice better than Deepgram. ## COMPANY OPERATING RHYTHM At Deepgram, we expect an AI-first mindset—AI use and comfort aren’t optional, they’re core to how we operate, innovate, and measure performance. Every team member who works at Deepgram is expected to actively use and experiment with advanced AI tools, and even build your own into your everyday work. We measure how effectively AI is applied to deliver results, and consistent, creative use of the latest AI capabilities is key to success here. Candidates should be comfortable adopting new models and modes quickly, integrating AI into their workflows, and continuously pushing the boundaries of what these technologies can do. Additionally, we move at the pace of AI. Change is rapid, and you can expect your day-to-day work to evolve just as quickly. This may not be the right role if you’re not excited to experiment, adapt, think on your feet, and learn constantly, or if you’re seeking something highly prescriptive with a traditional 9-to-5. ## THE OPPORTUNITY Voice is the most natural modality for human interaction with machines. However, current sequence modeling paradigms based on jointly scaling model and data cannot deliver voice AI capable of universal human interaction. The challenges are rooted in fundamental data problems posed by audio: real-world audio data is scarce and enormously diverse, spanning a vast space of voices, speaking styles, and acoustic conditions. Even if billions of hours of audio were accessible, its inherent high dimensionality creates computational and storage costs that make training and deployment prohibitively expensive at world scale. We believe that entirely new paradigms for audio AI are needed to overcome these challenges and make voice interaction accessible to everyone. ## THE ROLE Deepgram is currently looking for an experienced researcher to who has worked extensively with Large Language Models (LLMS) and has a deep understanding of transformer architecture to join our Research Staff. As a Member of the Research Staff, this individual should have extensive experience working on the hard technical aspects of LLMs, such as data curation, distributed large-scale training, optimization of transformer architecture, and Reinforcement Learning (RL) training. ## THE CHALLENGE We are seeking researchers who: - See "unsolved" problems as opportunities to pioneer entirely new approaches - Can identify the one critical experiment that will validate or kill an idea in days, not months - Have the vision to scale successful proofs-of-concept 100x - Are obsessed with using AI to automate and amplify your own impact If you find yourself energized rather than daunted by these expectations—if you're already thinking about five ideas to try while reading this—you might be the researcher we need. This role demands obsession with the problems, creativity in approach, and relentless drive toward elegant, scalable solutions. The technical challenges are immense, but the potential impact is transformative. ## WHAT YOU'LL DO - Brainstorming and collaborating with other members of the Research Staff to define new LLM research initiatives - Broad surveying of literature, evaluating, classifying, and distilling current methods - Designing and carrying out experimental programs for LLMs - Driving transformer (LLM) training jobs successfully on distributed compute infrastructure and deploying new models into production - Documenting and presenting results and complex technical concepts clearly for a target audience - Staying up to date with the latest advances in deep learning and LLMs, with a particular eye towards their implications and applications within our products ## YOU'LL LOVE THIS ROLE IF YOU - Are passionate about AI and excited about working on state of the art LLM research - Have an interest in producing and applying new science to help us develop and deploy large language models - Enjoy building from the ground up and love to create new systems. - Have strong communication skills and are able to translate complex concepts clearly - Are highly analytical and enjoy delving into detailed analyses when necessary ## IT'S IMPORTANT TO US THAT YOU HAVE - 3+ years of experience in applied deep learning research, with a solid understanding toward the applications and implications of different neural network types, architectures, and loss mechanism - Proven experience working with large language models (LLMs) - including experience with data curation, distributed large-scale training, optimization of transformer architecture, and RL Learning - Strong experience coding in Python and working with Pytorch - Experience with various transformer architectures (auto-regressive, sequence-to-sequence.etc) - Experience with distributed computing and large-scale data processing - Prior experience in conducting experimental programs and using results to optimize models ## IT WOULD BE GREAT IF YOU HAD - Deep understanding of transformers, causal LMs, and their underlying architecture - Understanding of distributed training and distributed inference schemes for LLMs - Familiarity with RLHF labeling and training pipelines - Up-to-date knowledge of recent LLM techniques and developments ## THE CHALLENGE We are seeking researchers who: - See "unsolved" problems as opportunities to pioneer entirely new approaches - Can identify the one critical experiment that will validate or kill an idea in days, not months - Have the vision to scale successful proofs-of-concept 100x - Are obsessed with using AI to automate and amplify your own impact If you find yourself energized rather than daunted by these expectations—if you're already thinking about five ideas to try while reading this—you might be the researcher we need. This role demands obsession with the problems, creativity in approach, and relentless drive toward elegant, scalable solutions. The technical challenges are immense, but the potential impact is transformative. ## WHAT YOU'LL DO - Brainstorming and collaborating with other members of the Research Staff to define new LLM research initiatives - Broad surveying of literature, evaluating, classifying, and distilling current methods - Designing and carrying out experimental programs for LLMs - Driving transformer (LLM) training jobs successfully on distributed compute infrastructure and deploying new models into production - Documenting and presenting results and complex technical concepts clearly for a target audience - Staying up to date with the latest advances in deep learning and LLMs, with a particular eye towards their implications and applications within our products ## YOU'LL LOVE THIS ROLE IF YOU - Are passionate about AI and excited about working on state of the art LLM research - Have an interest in producing and applying new science to help us develop and deploy large language models - Enjoy building from the ground up and love to create new systems. - Have strong communication skills and are able to translate complex concepts clearly - Are highly analytical and enjoy delving into detailed analyses when necessary ## IT'S IMPORTANT TO US THAT YOU HAVE - 3+ years of experience in applied deep learning research, with a solid understanding toward the applications and implications of different neural network types, architectures, and loss mechanism - Proven experience working with large language models (LLMs) - including experience with data curation, distributed large-scale training, optimization of transformer architecture, and RL Learning - Strong experience coding in Python and working with Pytorch - Experience with various transformer architectures (auto-regressive, sequence-to-sequence.etc) - Experience with distributed computing and large-scale data processing - Prior experience in conducting experimental programs and using results to optimize models ## IT WOULD BE GREAT IF YOU HAD - Deep understanding of transformers, causal LMs, and their underlying architecture - Understanding of distributed training and distributed inference schemes for LLMs - Familiarity with RLHF labeling and training pipelines - Up-to-date knowledge of recent LLM techniques and developments - Published papers in Deep Learning Research, particularly related to LLMs and deep neural networks - Published papers in Deep Learning Research, particularly related to LLMs and deep neural networks Notice: We're aware of individuals impersonating Deepgram recruiters. All legitimate Deepgram recruiting communication comes from an @deepgram.com http://deepgram.com email address. If you've received a message claiming to be Deepgram, please forward it to careers@deepgram.com. ## About Deepgram ## Company Overview - **One-liner**: Deepgram is a foundational AI company building real-time voice AI infrastructure for developers and enterprises to power speech-to-text, text-to-speech, and voice agent applications. - **Entity Type**: Private (Series C, Unicorn status) - **Headquarters**: San Francisco, California, USA (remote-first, distributed across 20+ states and 5+ countries) - **Founded**: 2015 - **Founders**: Scott Stephenson (CEO), Adam Sypniewski, Noah Shutty ## Core Business - **Primary industry**: Voice AI / Artificial Intelligence / Speech Recognition - **Target customers**: Developers, product teams, enterprises, and platforms embedding voice AI (B2B, B2D) - **Mission statement**: To transform the way humans interact with machines by unlocking the productivity of the world through voice technology ## Products & Services - **Speech-to-Text (STT)**: Real-time and batch transcription with models like Nova-3 (world’s most accurate real-time STT) and Flux Multilingual (first conversational STT in 10 languages). Available via cloud or self-hosted. [deepgram.com](https://deepgram.com/) - **Text-to-Speech (TTS)**: Aura-2, professional-grade enterprise TTS for natural voice output. [deepgram.com](https://deepgram.com/) - **Voice Agent API**: A unified API that combines STT, TTS, and LLM orchestration into a single endpoint, reducing complexity, latency, and cost for building conversational AI agents. [deepgram.com](https://deepgram.com/) - **Custom Models**: Ability to train custom speech models for unique use cases and domain-specific vocabulary. [deepgram.com](https://deepgram.com/) - **Saga (Voice OS)**: An operating system for voice applications, announced as a platform-level offering. [linkedin.com](https://www.linkedin.com/company/deepgram) - **Deepgram for Restaurants**: Vertical solution for drive-thru and restaurant voice ordering (95%+ containment), powered by the acquisition of OfOne. [linkedin.com](https://www.linkedin.com/company/deepgram) ## Market Standing - **Valuation**: $1.3 billion unicorn status (as of Series C in 2024) [deepgram.com](https://deepgram.com/about) - **Key Metric**: $130 million in Series C funding (total raised: ~$202M+ including $72M Series B) [deepgram.com](https://deepgram.com/about) - **Notable Investors/Partners**: Y Combinator, NVIDIA (proud partner), with Series C led by prominent venture firms [ycombinator.com](https://www.ycombinator.com/companies/deepgram) - **Growth Signals**: Processed 50,000+ years of audio and 1 trillion+ words; trusted by 200,000+ developers and 1,300+ organizations; acquired OfOne for restaurant voice AI; launched Flux Multilingual, Nova-3, Aura-2, and Saga in rapid succession [linkedin.com](https://www.linkedin.com/company/deepgram) ## Competitive Advantages - **Foundational AI moat**: Deepgram is a research-driven company that builds its own end-to-end deep learning models (not wrappers), giving them control over accuracy, latency, and cost. - **Unified API advantage**: A single Voice Agent API that replaces the need to stitch together separate STT, TTS, and LLM components—reducing complexity and latency. - **Scale and experience**: The most experienced voice AI platform in the world, having processed the largest volume of voice data (50,000+ years of audio). - **Enterprise-grade reliability**: Offers on-premise/self-hosted deployment options for compliance-heavy industries (healthcare, finance, legal). - **NVIDIA partnership**: Access to scalable GPU infrastructure for training and inference, differentiating them from competitors. ## Strategic Focus - **Voice AI Economy leadership**: Positioning Deepgram as the infrastructure layer for the emerging "trillion-dollar Voice AI economy." - **Multilingual expansion**: Launching Flux Multilingual to capture global markets beyond English. - **Vertical industry penetration**: Moving beyond horizontal API into vertical solutions (restaurants via OfOne acquisition, medical transcription). - **Agentic AI**: Doubling down on Voice Agent API and autonomous voice agents as the next growth vector. - **Platform expansion**: Building Saga (Voice OS) to become the operating system for all voice applications. ## Why Work Here - **Remote-first culture**: Deepgram is a remote-first company with employees across 20+ states in the US and 5+ countries. They offer freedom and flexibility to work from anywhere. [deepgram.com](https://deepgram.com/careers) - **Strong benefits package**: Medical, dental, vision insurance; mental health support; unlimited PTO; 12 paid US holidays; parental leave; 401(k) with company match; wellness stipend; learning/education stipend; quarterly productivity stipend; one-time office upgrade stipend. [deepgram.com](https://deepgram.com/careers) - **Inclusive environment**: Active Employee Resource Groups (ERGs) and a stated commitment to "every voice, heard and understood." [deepgram.com](https://deepgram.com/careers) - **Annual offsites**: While remote-first, the company convenes annually for in-person offsites to foster collaboration and connection. [deepgram.com](https://deepgram.com/careers) - **Engineering culture**: Values include "be curious," "grow together," "do the right thing," "think smart, act fast," and "put the customer first." The culture is described as self-motivated, positive, passionate, and competitive. [deepgram.com](https://deepgram.com/careers) - **Research-driven environment**: Deepgram was founded by physicists and maintains a strong research team (Research Scientist, Research Engineer roles), offering the chance to work on cutting-edge AI. [ycombinator.com](https://www.ycombinator.com/companies/deepgram) - **Salary transparency**: Publicly lists salary ranges for roles (e.g., Research Scientist $150K-$250K, SDR $50K-$60K), indicating compensation transparency. [ycombinator.com](https://www.ycombinator.com/companies/deepgram) ## Sources 1. [deepgram.com - Careers Page](https://deepgram.com/careers) 2. [deepgram.com - Product Page](https://deepgram.com/) 3. [deepgram.com - About Page](https://deepgram.com/about) 4. [ycombinator.com - Deepgram Profile](https://www.ycombinator.com/companies/deepgram) 5. [linkedin.com - Deepgram Company Page](https://www.linkedin.com/company/deepgram) ## Other roles at Deepgram - [Talent Acquisition Manager](https://feeny.ai/job/talent-acquisition-manager-deepgram-united-states-37a7eh07z31n) — United States - [Strategic Account Execustive (New York City)](https://feeny.ai/job/strategic-account-execustive-new-york-city-deepgram-new-york-pddfzv4pr2rs) — New York, NY - [Executive Business Parter, Customer Success](https://feeny.ai/job/executive-business-parter-customer-success-deepgram-united-states-v8qhnkgewtt2) — United States - [Senior Forward Deployed Engineer (FDE), Strategic Accounts](https://feeny.ai/job/senior-forward-deployed-engineer-fde-strategic-accounts-deepgram-new-york-g66c10a7xrb5) — New York, NY - [Software Test Engineer](https://feeny.ai/job/software-test-engineer-deepgram-united-states-dhq28zwk1dr0) — United States - [GM, Sales (EMEA)](https://feeny.ai/job/gm-sales-emea-deepgram-london-9f79v8a9f0bf) — London, United Kingdom - [Partner Success Manager (AWS)](https://feeny.ai/job/partner-success-manager-aws-deepgram-united-states-k7yatsds8fc5) — United States - [Revenue Accounting Manager](https://feeny.ai/job/revenue-accounting-manager-deepgram-united-states-5nwxqr5ps70y) — United States - [Account Executive (Melbourne)](https://feeny.ai/job/account-executive-melbourne-deepgram-sydney-t76trcyj8s14) — Sydney, Australia - [Research Staff, Data Science](https://feeny.ai/job/research-staff-data-science-deepgram-united-states-pf9zk68t5qs2) — United States