Backend Engineer- Inference Services
Deepgram · Remote · USA - Remote ·
COMPENSATION
$150k – $220k
Listed on the employer's job board (Ashby)
- Fully remote
- Rust, C++, Python
- 3+ years experience
- Voice AI platform
- Systems optimization
Join the Engine team at Deepgram to design and implement secure, robust services for real-time speech-to-text and text-to-speech APIs. You will work on efficient distributed compute orchestration and optimized scheduling to support production-grade voice agents at scale.
The company operates with an AI-first mindset, expecting team members to actively integrate advanced AI tools into their daily workflows. Success in this role requires adapting quickly to rapid changes and continuously experimenting with new technologies.
What you'll do
- Improve core inference services including networking and audio transcoding
- Optimize latency and memory usage for high-performance systems
- Develop processes to measure and maximize system performance
- Debug complex issues involving scheduling and high-performance computing
- Customize backend services rapidly to meet customer needs
- Partner with Product to design and implement features end-to-end
What you'll need
- At least three years of industry experience
- Proficiency in Rust, C, or C++
- Competence in Python programming
- Strong written and verbal communication skills
- High-level experience with version control, preferably git
- Comprehensive experience with UNIX-style systems
Nice to have
- Experience with modern machine learning frameworks like Torch
- Knowledge of CNN, RNN, and transformer architectures
- Background in audio processing
Stack
- Rust
- C
- C++
- Python
- Git
- UNIX
- Torch
How you'll work
- Fully remote position
- Fast-paced environment with rapid change
- Requires active use of AI tools in daily work