Member of Technical Staff

About the role

Perplexity is seeking energetic and talented engineers to join our highly driven Agents engineering team. This team comprises backend, full-stack, and AI/ML engineers who work collaboratively to develop harnesses and AI systems that power engaging and agentic experiences for users. As a member of this team, you will be instrumental in building and refining systems that enable our products to deliver generalized frontier intelligence, such as the Perplexity Computer platform, the Comet ecosystem, and our Agent API. Your work will directly contribute to our vision of empowering users with agents capable of faithfully actualizing their intent through open-ended interactions across various environments. You will be involved in both applied research and engineering, tackling open problems in AI, including designing intelligent agents, training decision models, ensuring user experiences, and developing secure architectures. This role offers a unique opportunity to work at the forefront of AI technology, shaping products that serve millions of dedicated users.

Key facts

What you'll do

• Engineer agent harnesses that connect powerful AI models with environments and tools necessary to perform valuable and economically beneficial tasks for users.
• Drive the development of AI capabilities across multiple layers of a rapidly growing product focused on AI agents.
• Build and optimize AI models, infrastructure, and browser technologies to push the capability frontier and scale those capabilities for a large and expanding user base.
• Ensure high standards of craftsmanship and quality in AI agent performance, user experience, and overall reliability.
• Collaborate closely with fellow engineers, designers, product managers, data scientists, and other stakeholders to integrate core functionalities into frontier agentic products and ensure operation.
• Contribute to maintaining product reliability, code quality, testing, and ongoing system maintenance across the broader engineering team.
• Design and implement secure permission architectures, payload classifiers, and other mechanisms to enable secure-by-design agent capabilities.
• Develop data representations and modes of interaction between agents and their environments to optimize performance and user engagement.
• Address open research and engineering challenges related to AI navigation, multimodal decision-making, and environment interaction.
• Participate in cross-disciplinary efforts to advance the frontier of AI agent technology, including experimenting with new models, architectures, and interaction paradigms.
• Work on integrating AI models with browser technologies such as Chrome DevTools Protocol (CDP), Playwright, and extension development to enhance agent capabilities across platforms.
• Contribute to the development of tools and frameworks that facilitate the training, evaluation, and deployment of multimodal AI models.
• Engage in continuous learning to stay updated with the latest advancements in AI, ML, and related technologies, applying them to improve product features.
• Support efforts to implement secure and privacy-preserving features, ensuring compliance with relevant standards and best practices.
• Take ownership of projects from conception through deployment, demonstrating initiative and a proactive approach to problem-solving.
• Participate in code reviews, knowledge sharing, and mentoring within the team to foster a collaborative and high-performance engineering culture.

Requirements

• Strong foundational understanding of the full AI product stack, including model training, deployment, and inference.
• Proficiency in Python, with experience applying it to AI and ML workflows.
• Bonus points for experience with TypeScript, Go, and Rust, especially in the context of AI or browser technologies.
• Context engineering and tool interfaces for frontier AI models.
• Post-training and reinforcement learning, particularly for multimodal models.
• Browser technologies such as Chrome DevTools Protocol (CDP), Playwright, extension development, or related areas.
• Strong product intuition and a keen sense for user experience excellence.
• Ability to work effectively within a small, fast-moving team, taking ownership of projects and driving them to completion.
• Passion for shipping innovative products that surprise and delight users, with a focus on practical impact and usability.
• Experience working on AI systems involving multimodal models, complex decision-making, and environment interaction.
• Familiarity with secure architecture design principles, especially for agent capabilities.
• Excellent communication skills and the ability to collaborate across disciplines.
• Willingness to dive into complex problems and contribute to both research and engineering efforts.
• Demonstrated ability to adapt to rapidly evolving technology landscapes and incorporate new knowledge into work.

Nice to have

• Experience with open-ended AI interactions and frontier AI models.
• Background in applied research related to AI and machine learning.
• Familiarity with cloud infrastructure, deployment, and scaling strategies.
• Prior experience with AI evaluation, testing methodologies, and performance benchmarking.
• Knowledge of privacy and security considerations in AI systems.
• Exposure to the latest advancements in AI, including multimodal and reinforcement learning techniques.

Skills & tools

• Python for AI development and scripting.
• TypeScript for frontend or tooling development (bonus).
• Go and Rust for systems programming and performance-critical components (bonus).
• Browser technologies such as Chrome DevTools Protocol (CDP), Playwright, and extension development.
• AI and ML frameworks and models relevant to multimodal and reinforcement learning.
• Infrastructure and backend development skills for deploying scalable AI systems.
• Security and permission architecture design to ensure safe and compliant agent capabilities.
• Version control systems, testing frameworks, and CI/CD pipelines.

Practical notes

• This is an on-site role based in San Francisco, requiring presence at the office.
• The position involves working closely with a multidisciplinary team to develop, test, and deploy AI agents and systems.
• Candidates should be comfortable working in a dynamic environment, taking ownership of projects, and iterating quickly.
• The role offers an to contribute to frontier AI research and engineering, impacting products used by millions of users worldwide.
• You will be expected to stay current with emerging AI technologies and incorporate new techniques into product development.
• The work involves integrating AI models with browser and platform technologies to create agent experiences.
• Candidates should be proactive, collaborative, and eager to solve complex open problems in AI and system design.