Note: The job is a remote job and is open to candidates in USA. Deepgram is the leading platform for Voice AI, providing real-time APIs for speech-to-text and text-to-speech services. They are seeking a Software Engineering Intern for Fall 2026/Summer 2027 to design, build, and ship a scoped project while contributing to the production Deepgram codebase and exploring voice AI technologies.
Responsibilities
- Design, build, and ship one scoped project end to end, from design through reviewed, tested code running in staging or production, with a dedicated mentor guiding you at each milestone
- Contribute directly to a production Deepgram codebase, whether that's the core voice AI platform, the Applied AI wing (Deepgram for Restaurants), or the consumer wing, landing merged PRs that teammates and customers actually use
- Dig into voice AI: speech and audio ML, real-time systems, and how research, engineering, and customers form one feedback loop
- Use Agentic tooling (Claude Code, Codex, whatever you want!!) as a default part of how you prototype, test, and debug, and bring at least one workflow improvement back to the team
Skills
- You've built things because you wanted them to exist: projects, tools, scripts, or automations, whether in class, on your own, or in a prior role
- You reach for AI as a default part of how you learn and build, not an occasional add-on, and you can talk about where it helps and where human judgment still has to lead
- You reason from first principles: when something breaks, you dig into why rather than patching around it
- You write and read code in at least one language, and you pick up new languages, tools, and codebases quickly
- You can explain your work clearly: what you built, what broke, and what you'd do differently
- You treat 'good enough' as a question, not a finish line, and you're drawn to hard problems
- You give and receive feedback well and want to get better fast
- Currently pursuing a degree in computer science, engineering, or a related field, or building equivalent skills through self-study, open source, or your own projects
- Coursework or hands-on exposure to machine learning, real-time systems, or audio/speech processing
- A prior internship, a hackathon project, or something you built and shipped for yourself, ideally with an AI-assisted workflow behind it
Company Overview