- Job title
- Senior Software Engineer (CVI)
- Job location
- San Francisco
- Job description
- ## **About Us**
[Tavus](https://www.tavus.io/) is a research lab pioneering human computing. We’re building AI Humans: a new interface that closes the gap between people and machines, free from the friction of today’s systems. Our real-time human simulation models let machines see, hear, respond, and even look real—enabling meaningful, face-to-face conversations. AI Humans combine the emotional intelligence of humans with the reach and reliability of machines, making them capable, trusted agents available 24/7, in every language, on our terms.
Imagine a therapist anyone can afford. A personal trainer that adapts to your schedule. A fleet of medical assistants that can give every patient the attention they need. With Tavus, individuals, enterprises, and developers can all build AI Humans to connect, understand, and act with empathy at scale.
We’re a Series B company backed by world-class investors including **Sequoia Capital, Y Combinator, and Scale Venture Partners**.
Be part of shaping a future where humans and machines truly understand each other.
## **The Role**
We're hiring a Senior Software Engineer to be a technical driver on Tavus's Conversational Video Interface, our multimodal real-time conversational product. It's a hands-on role. Your code runs in the middle of live conversations between people and PALs.
## **What you'll own:**
- **Ship features end-to-end.** You'll take product priorities that start out loosely defined, like voice localization for accent support or multilingual sentence endpointing, and turn them into a spec, an architecture, and shipped features.
- **Partner with Research** to bring state-of-the-art real-time models into CVI. Sometimes that means dipping into research territory yourself.
- **Fix what you find.** If you spot a flaky process or a slow path, you don't need a ticket or anyone's permission to go after it.
## **What this role has shipped:**
- **Guided Browser Control** : a framework that lets PALs drive a real web browser mid-conversation, narrating each click and scroll out loud
- **Magic Canvas** : interactive cards rendered right on the video stream. The PAL puts the right card up at the right moment: a question, a text box, a calendar, a chart, an alert
- **Memories and Knowledge Base** : the context layer that lets PALs recall past conversations and pull from document stores in real time, without breaking stride
- **Utterance-to-utterance optimization** : shaving hundreds of milliseconds off conversational turns
## **Who you are:**
- **You're proactive.** When you see an issue, you fix it. Nobody has to ask.
- **You're autonomous.** Given a loosely defined project, you figure out what "done" means, make the judgment calls, and deliver.
- **You're energized by unfamiliar problems.** A lot of the work here has no playbook, and you'll take on projects outside your direct experience.
- **You adapt as priorities evolve.** This space moves fast, and the most important thing to build can change as we learn. When it does, you adjust and keep moving.
- **You hold a high bar.** What matters to you is how the product feels to the person on the other end of the conversation. Clean code and clever architecture are means to that end.
- **You're passionate about the future of interfaces.** How people and computers interact is being reinvented right now, and you want a hand in shaping it.
## **Requirements:**
- **Fluent in Python, with real depth in concurrent programs.** Low-level concepts (asyncio, multiprocessing, event loops, the GIL, shared memory) are dear friends, and you design with them in mind.
- **You've built real-time systems** , whether video, audio, or model inference, and you know what a millisecond latency budget feels like.
- **You bring a senior track record** : you set direction, own decisions, and get things over the finish line.
- **You explain complex things clearly** , and you use that to teach the people around you.
## **Nice to haves:**
None of these are required, but they'd help.
- Experience with GPUs and GPU management
- Deep experience with video streaming over WebRTC
- Deep experience with LLMs and agentic frameworks
- Extensive experience with low-level systems
If you don't check every box but this sounds like the work you want to be doing, apply anyway.
