Senior Software Engineer (CVI)

Tavus • San Francisco, CA

6d

About The Position

Tavus is a research lab pioneering human computing, building AI Humans that act as a new interface between people and machines. These AI Humans enable meaningful, face-to-face conversations by simulating real-time human interaction, combining emotional intelligence with the reach and reliability of machines. Tavus is a Series B company backed by prominent investors and is seeking individuals passionate about shaping the future of human-computer interaction.

The role of Senior Software Engineer focuses on the Conversational Video Interface (CVI), Tavus's multimodal real-time conversational product. This is a hands-on position where the engineer's code will be integral to live conversations between people and AI Humans (PALs). The engineer will be responsible for shipping features end-to-end, from loosely defined product priorities to architecture and implementation. This includes tasks like voice localization for accent support and multilingual sentence endpointing. The role involves partnering with the Research team to integrate state-of-the-art real-time models into CVI and proactively addressing system improvements, such as optimizing slow paths or flaky processes. Past shipped features include Guided Browser Control, Magic Canvas for interactive elements on video streams, Memories and Knowledge Base for context recall, and utterance-to-utterance optimization for reduced latency.

Requirements

• Fluent in Python, with deep understanding of concurrent programming concepts (asyncio, multiprocessing, event loops, the GIL, shared memory).
• Experience building real-time systems (video, audio, or model inference) with an understanding of millisecond latency budgets.
• A senior track record demonstrating the ability to set direction, own decisions, and deliver projects to completion.
• Ability to explain complex things clearly and teach people around you.

Nice To Haves

• 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.

Responsibilities

• Ship features end-to-end, taking loosely defined product priorities and turning them into a spec, an architecture, and shipped features.
• Partner with Research to bring state-of-the-art real-time models into CVI.
• Proactively identify and fix system issues, such as flaky processes or slow paths, without requiring explicit tickets or permission.
• Take ownership of technical direction and decisions for projects.
• Clearly explain complex concepts and teach others.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed