Sr. Staff Software Engineer, Pinterest Assistant

About the role

You will own the end-to-end technical strategy and execution for the backend systems that power Pinterest Assistant, ensuring the platform reliably serves millions of users with high performance and scalability. You will define the architecture for multi-modal, conversational experiences and lead cross-functional collaboration to turn product vision into production-grade features. This role requires you to balance rapid experimentation with long-term maintainability while establishing engineering standards for reliability and safety in AI-driven systems. You will mentor engineers, drive technical excellence, and act as a trusted advisor to product and design teams. Your work will directly impact how Pinners discover inspiration and take action in their daily lives. You will navigate ambiguity and make pragmatic decisions that align with business goals and user trust.

Key facts

What you'll do

Define and drive the technical vision and architecture for the backend systems powering Pinterest Assistant, ensuring they scale reliably as the product reaches millions of users.

Lead the design and implementation of the core Assistant platform including orchestration, memory and personalization, context management, APIs, tool integrations, and backend infrastructure.

Drive technical strategy, identifying foundational platform investments that unlock new Assistant capabilities while improving developer velocity, reliability, and scalability.

Partner closely with Product, Design, Frontend engineers, ML engineers, and Infrastructure teams to build a deeply personalized, multimodal conversational experience.

Shape the architecture for AI-powered applications, including retrieval, context assembly, tool execution, memory, and model orchestration, in partnership with ML teams.

Establish engineering best practices for production LLM and agent systems, including observability, logging, evals, safety mechanisms, rollout strategies, and incident response.

Mentor engineers, raising the technical bar through architecture reviews, technical guidance, and thoughtful engineering leadership.

Navigate ambiguity and make pragmatic architectural decisions that balance rapid product iteration with long-term maintainability.

Influence technical direction beyond the immediate team by collaborating across teams and helping shape Pinterest's broader AI platform strategy.

Stay current with advances in LLMs, agentic systems, and backend platform design, and apply them pragmatically to consumer product development.

Own key reliability and performance metrics for Assistant services, driving improvements and ensuring a high-quality user experience at scale.

Collaborate with SRE and platform teams to ensure robust deployment pipelines, monitoring, and operational excellence for backend services.

Evaluate and integrate emerging technologies and third-party tools to enhance Assistant capabilities while maintaining security and compliance standards.

Communicate progress, risks, and trade-offs clearly to stakeholders, aligning technical decisions with business outcomes and user needs.

Requirements

8+ years of backend engineering experience with a strong track record of delivering scalable, reliable production systems.

Proficiency in one or more backend programming languages such as Java, Python, Go, or Node.js, with demonstrated ability to write clean, maintainable, and performant code.

Deep understanding of distributed systems principles, including concurrency, fault tolerance, observability, and capacity planning.

Experience designing and operating high-throughput, low-latency APIs and services in cloud environments.

Strong knowledge of database systems, including relational and NoSQL stores, caching strategies, and data modeling for scale.

Experience with machine learning platforms, model serving infrastructure, or AI/agent systems is required.

Solid understanding of security best practices, authentication, authorization, and data privacy in consumer-facing products.

Ability to lead technical discussions and influence architecture across multiple teams without direct authority.

Nice to have

Experience with large-scale language models, retrieval-augmented generation, or agent orchestration frameworks.

Background in building recommendation, search, or personalization systems.

Contributions to open source projects or technical talks at industry conferences.

Practical notes

This is a full-time role located in San Francisco, CA, US, or remote within the United States.

Visa sponsorship is available for eligible positions.

Hours are full-time, during standard business hours with flexibility for asynchronous work.

Travel is not required for this role.

Applications will be reviewed on a rolling basis until the position is filled.