About this role
Scaling realtime voice and vision infrastructure requires overcoming significant latency, reliability, and streaming challenges as users increasingly expect agentic AI to take action rather than just output text. Perplexity needs an Engineering Manager to lead the Multimodal team, guiding the architecture that connects live audio sessions to frontier models and backend tools. This group owns everything from client-facing SDKs to server-side orchestration, bridging the gap between raw research models and daily consumer applications.
Reporting to this position involves setting technical direction, establishing product roadmaps, and steering new features from early prototypes to production launches. The person stepping into this role will hire, mentor, and develop talented engineers while maintaining high execution standards in a fast-moving environment. Ideal candidates bring a strong background in backend systems, proven management experience, and a genuine curiosity for emerging artificial intelligence applications.
What you will do
- Lead the team building realtime voice and multimodal experiences across the platform from prototype to production.
- Own the product roadmap by determining which voice and vision features and underlying infrastructure to prioritize next.
- Hire, coach, and develop software engineers while granting them ownership of complex technical challenges.
- Shape the architectural design for live audio interactions from user input to model responses and agent execution.
- Guide engineering efforts through strict latency, reliability, and scaling hurdles associated with live voice sessions.
- Collaborate closely with client, infrastructure, and model teams to integrate new audio and vision capabilities into applications.
- Leverage production performance metrics and user feedback to refine features and guide future investment.
What you need
- Demonstrated experience managing engineering teams through hiring, coaching, and performance management.
- A solid software engineering foundation centered on backend or distributed systems.
- Proven history of designing and operating production services within rapidly scaling cloud environments like AWS.
- A track record of steering teams to successfully ship complex projects in fast-paced settings.
- Strong product judgment with the ability to turn user pain points into actionable technical priorities.
- Clear communication skills enabling effective collaboration across multiple cross-functional teams.
- Familiarity with programming languages such as Rust, Python, or Go, with primary work done in Rust and Python.
- A genuine interest in adopting AI products alongside a willingness to learn quickly.
Nice to have
- Experience with realtime media systems including voice, audio streaming, WebRTC, or low-latency transport.
- Experience integrating LLMs, speech models, or computer vision into production systems.
- Experience with agent frameworks, tool-calling architectures, or sandboxed execution environments.
- Experience leading projects spanning backend services, client SDKs, and user-facing applications.
- Time spent leading a team at a fast-growing startup or in a high-ownership environment.
Benefits
- Offers Equity
Skills and tech stack
Questions candidates ask
What is the compensation range for this role?
The salary for this position is $300K to $405K with offered equity.
Where is the job located?
The position is located in San Francisco.
What programming languages does the team use?
The team works primarily in Rust and Python, while also considering experience with Go and similar languages.
What does the Multimodal team actually build?
The team builds the experiences and infrastructure that move AI interaction beyond touch and text into realtime voice, vision, and platform systems.