Senior Research Engineer at AssemblyAI (S17) $210K - $309K The best way to build Voice AI apps Remote / Remote (US) Full-time US citizen/visa only 3+ years About AssemblyAI AssemblyAI is the #1 rated API for Speech Recognition. Thousands of developers use our APIs to transcribe millions of videos, podcasts, phone calls, and zoom meetings every day - powering innovative products like visual voicemail, meeting summarizers, and closed captioning. We're backed by leading investors including Y Combinator, John and Patrick Collison of Stripe, Daniel Gross, and Nat Friedman. We were founded in 2017. About the role Skills: Torch/PyTorch, Machine Learning, Deep Learning About AssemblyAI At AssemblyAI, we’re building at the forefront of Speech AI, creating powerful models for speech-to-text and speech understanding available through a straightforward API. With more than 200,000 developers building on our API and over 5,000 paying customers, AssemblyAI is helping unlock and support the next generation of powerful, meaningful products built with AI. Progress in AI is moving at an unprecedented pace– and our team is made up of experts in AI research that are focused on making sure that our customers are able to stay on the cutting edge, with production-ready AI models that are constantly updating and improving as our team continues to improve accuracy, latency, and what’s possible with Speech AI. Our models consistently rank highest in industry benchmarks for accuracy, outperforming models from Google and Amazon, and up to 30% fewer hallucinations than OpenAI’s Whisper. Our models power more than 2 billion end-user experiences each day, helping companies better understand customer feedback, run more productive meetings with automated meeting notes, and helping improve childhood literacy via ed tech tools. We’ve raised funding by leading investors including Accel, Insight Partners, Y Combinator’s AI Fund, Patrick and John Collision, Nat Friedman, and Daniel Gross. We’re a remote team looking to build one of the next great AI companies, and are looking for driven, talented people to help us get there! About the Role We are seeking a highly skilled Senior Research Engineer to collaborate closely with both Research and Engineering teams. The role involves diagnosing and resolving bottlenecks across large-scale distributed training, data processing, and inference systems, while also driving optimizations for existing high-performance pipelines. The ideal candidate possesses a deep understanding of modern deep learning systems, combined with strong engineering expertise in areas such as layer-level optimization, large-scale distributed training, streaming, low-latency and asynchronous inference, inference compilers, and advanced parallelization techniques. This is a cross-functional role requiring strong technical rigor, attention to detail, intellectual curiosity, and excellent communication skills. The position is embedded within the Research team and is responsible for developing and refining the technical foundation that enables cutting-edge research and translates its outcomes into production, bridging research and production engineering. What You'll Do Investigate and mitigate performance bottlenecks in large-scale distributed training and inference systems. Develop and implement both low-level (operator/kernel) and high-level (system/architecture) optimization strategies. Translate research models and prototypes into highly optimized, production-ready inference systems. Explore and integrate inference compilers such as TensorRT, ONNX Runtime, AWS Neuron and Inferentia, or similar technologies. Design, test, and deploy scalable solutions for parallel and distributed workloads on heterogeneous hardware. Facilitate knowledge transfer and bidirectional support between Research and Engineering teams, ensuring alignment of priorities and solutions. What You'll Need Strong expertise in the Python ecosystem and major ML frameworks (PyTorch, JAX). Experience with lower-level programming (C++ or Rust preferred). Deep understanding of GPU acceleration (CUDA, profiling, kernel-level optimization); TPU experience is a strong plus. Proven ability to accelerate deep learning workloads using compiler frameworks, graph optimizations, and parallelization strategies. Solid understanding of the deep learning lifecycle: model design, large-scale training, data processing pipelines, and inference deployment. Strong debugging, profiling, and optimization skills in large-scale distributed environments. Excellent communication and collaboration skills, with the ability to clearly prioritize and articulate impact-driven technical solutions. Pay Transparency: AssemblyAI strives to recruit and retain exceptional talent from diverse backgrounds while ensuring pay equity across our team. Our salary ranges are set to be competitive for our size, stage, and industry, and reflect just one component of the full compensation, benefits, and rewards we offer. Salary determinations consider a variety of factors, including relevant experience, technical depth, skills demonstrated during the interview process, and maintaining internal equity with peers on the team. The range shared below represents a general expectation for the posted position. However, we are open to considering candidates who may fall above or below the outlined experience level—in those cases, we will communicate any adjustments to the expected salary range. The range provided applies to candidates located in the United States. For candidates outside of the U.S., compensation ranges may differ; any adjustments will be communicated throughout the interview process. Salary range: $210,000 - $309,000 The expected base compensation for this role is listed above. Our total compensation package includes competitive equity grants, 100% employer-paid benefits, and the flexibility of being fully remote. Working at AssemblyAI We are a small but mighty group of startup veterans and experienced AI researchers with over 20 years of expertise in Machine Learning, Speech Recognition, and NLP. As a fully remote team, we’re looking for people to join our team who are ambitious, curious, and lead with integrity. We’re still in the early days of AI and of AssemblyAI’s journey, and are looking for teammates who won’t just fit in, but will help us define and build our company culture. We’re committed to creating a space where our employees can bring their full selves to work and have equal opportunity to succeed. No matter your race, gender identity or expression, sexual orientation, religion, origin, ability, age, veteran status, if joining this mission speaks to you, we encourage you to apply! Using AI to Interview: If you’re selected for an interview, please review this resource to better understand how AssemblyAI approaches the use of AI in our interview process. Keep Exploring AssemblyAI: Check us out on YouTube! Learn more about AI models for speech recognition Core Transcription | Audio Intelligence | LeMUR | Try the Playground Our $50M Series C fundraise Technology All of our Deep Learning models are trained in PyTorch on extremely large in-house datasets, across dozens of v100s, in our dedicated training clusters. For our production stack, we deploy everything into a modern Python micro-service architecture in AWS (Flask, Tornado, Docker, etc).