Software Engineer

Software Engineer to design, develop, test, and maintain applications, focusing on LiveKit Agents framework with Python and Node.js.

LiveKit is building the infrastructure layer for the voice-driven era of computing. Our platform gives developers everything they need to build, test, deploy, scale, and observe agents in production. Founded in 2021, LiveKit powers voice AI applications for OpenAI, xAI, Salesforce, Coursera, Spotify, and thousands of others, collectively facilitating billions of calls each year.

You'll Thrive at LiveKit if you:

obsess with crafting code that is fast, reliable and practical for the problem
are known as the go-to person for tackling tough technical problems
work hard and can build and ship fast
can clearly explain complex technical concepts to others
are a fast learner, frequently picking up new languages and tools

The best way to impress us is with thoughtful Issues and/or PRs on our Github repos 😊

What You’ll Do:

Design, development, testing, and maintenance of software applications.
Design and develop core functionalities for the LiveKit Agents framework with Python and Node.js.
Improve the foundational framework for building voice-driven Large Language Model (LLM)-based applications.
Build framework plugins that integrate SOTA Speech-To-Text (STT)/ Text-To-Speech (TTS)/LLM models from third-party providers.
Write clear documentation and examples for users.
Work with WebRTC, Voice Activity Detection (VAD), Semantic Turn Detector Model (EOU), Voice Pipeline, Multi-agent orchestration, and Model Context Protocol (MCP).

Who You Are:

You have a Bachelor of Science degree in Computer Science & Mathematics, Computer Science, Computer Engineering, or a closely related field.
You have least 18 months of experience in the job offered or at least 18 months of experience in the following:

Developing scalable generative AI and machine learning infrastructures
Speech-related model optimizations, including setting up voice pipelines using WebRTC, VAD, and a semantic turn detector model
Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO) for LLM models
Multi-agent architecture: ReAct, Chain-of-Thought, Tree-of-Thought
Programming languages: Python & Lua
Retrieval-Augmented Generation (RAG) and Vector databases for natural language processing

Our Commitment to You:

An opportunity to build something truly impactful to the world
Contribute to open source alongside world-class engineers
Competitive salary and equity package
Health, dental, and vision benefits
Flexible vacation policy
Various worksites—job may allow for remote work.

You'll Thrive at LiveKit if you:

obsess with crafting code that is fast, reliable and practical for the problem

are known as the go-to person for tackling tough technical problems

work hard and can build and ship fast

can clearly explain complex technical concepts to others

are a fast learner, frequently picking up new languages and tools

The best way to impress us is with thoughtful Issues and/or PRs on our Github repos 😊

What You’ll Do:

Design, development, testing, and maintenance of software applications.

Design and develop core functionalities for the LiveKit Agents framework with Python and Node.js.

Improve the foundational framework for building voice-driven Large Language Model (LLM)-based applications.

Build framework plugins that integrate SOTA Speech-To-Text (STT)/ Text-To-Speech (TTS)/LLM models from third-party providers.

Write clear documentation and examples for users.

Work with WebRTC, Voice Activity Detection (VAD), Semantic Turn Detector Model (EOU), Voice Pipeline, Multi-agent orchestration, and Model Context Protocol (MCP).

Who You Are:

You have a Bachelor of Science degree in Computer Science & Mathematics, Computer Science, Computer Engineering, or a closely related field.

You have least 18 months of experience in the job offered or at least 18 months of experience in the following:

Developing scalable generative AI and machine learning infrastructures

Speech-related model optimizations, including setting up voice pipelines using WebRTC, VAD, and a semantic turn detector model

Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO) for LLM models

Multi-agent architecture: ReAct, Chain-of-Thought, Tree-of-Thought

Programming languages: Python & Lua

Retrieval-Augmented Generation (RAG) and Vector databases for natural language processing