Deepgram
Backend Engineer- Inference Services
About this role
Deepgram seeks a Backend Engineer to build and optimize core inference services powering their voice AI platform. You'll design scalable, secure services for speech processing, distributed compute orchestration, and system performance optimization while working in a fast-paced, AI-driven environment.
What you'll do
- Improve core inference services including networking, speech processing, audio transcoding, and latency/memory optimization
- Develop measurement and optimization processes to maximize system performance
- Debug complex system issues spanning networking, scheduling, and high-performance computing
- Customize backend services rapidly to support customer requirements
- Partner with Product to design and implement new services and features end-to-end
What they're looking for
- Rust (or C/C++) programming
- Python development
- UNIX/Linux systems administration
- Git and version control
- Distributed systems and compute orchestration
- Audio processing (preferred)
- Machine learning frameworks or architectures (preferred)
- Performance optimization and debugging
Benefits
- Remote work opportunity
- Impact-driven role at a well-funded Series C company
- Work on trillion-dollar Voice AI economy infrastructure
- Continuous learning environment with emerging technologies
- Collaborative culture with product and engineering teams
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Deepgram
Deepgram builds voice AI and speech processing technology, offering platforms that power real-time speech recognition and audio intelligence across cloud, edge, and embedded environments. The company is hiring applied ML engineers, backend engineers, embedded AI engineers, and customer-facing roles (customer success and pre-sales engineers) to scale its AI infrastructure and drive enterprise adoption.
- Website
- deepgram.com
Likely interview questions
- Can you walk us through a complex system issue you debugged that involved networking or scheduling, and how you approached it?
- How would you balance shipping a minimal viable change versus investing time in detailed design work for a new service?