Staff Software Engineer, Plaid

• Strong SWE fundamentals (system design, APIs, etc.)
• Hands-on experience working with LLMs to build products and shipping them to product with iterating with real user feedback - including but not limited to:
• Prompt engineering
• Fine-tuning
• Retrieval augmented generation (RAG)
• Semantic search - vector database and embedding models
• Agent orchestration framework
• Evaluation and monitoring framework of open-ended tasks
• Streaming and SSE
• Common UX and design patterns for GenAI-powered products
• Strong debugging and monitoring experience for production systems
• Ability to deeply understand customer and user needs through user research and rapid experimentation - be your own technical PM
• Ability to balance divergent thinking (exploring possibilities) with convergent thinking (evaluating feasibility), which is critical for driving 0 ->1 projects
• Extremely curious and passionate about working in GenAI applications space

Desirable

• You will have the opportunity to join as one of the founding members of this newly formed team that is dedicated to consolidating and rapidly scaling our successful bets so far, and grow with the team in our quest to accelerate Plaid’s transformation into an AI-first company
• In this role you will lead projects that enable and scale our business with our largest AI customers and partners, starting with personal finance use cases and expanding into many others; examples include:
• Develop and evolve the preferred integration pattern for Plaid with AI providers - from API adaptations to building the official Plaid MCP Servers, and beyond
• Redefine how Plaid’s consumer link experience embed into conversational interfaces in the most seamless way
• Architect the trust layer for the future of agentic commerce that will become the industry standard
• Additionally you will be expected to scale and extend our existing successful bets on AI-powered customer experience; examples include:
• Make the next step-function improvement in our homegrown customer support agent - land our multi-turn and multi-agent system that powers a truly delightful experience for our customers; define how to scalably run offline evaluation for complex multi-turn open-ended tasks; research and prototype how Human-In-The-Loop - Reinforcement Learning (RLHF) can power an insights flywheel; pioneer the architecture for customer-specific long-term memory, etc
• Extend our agentic system to support other critical parts of the customer journey, starting with areas with the highest ROI - top-of-funnel product recommendation, customer onboarding and risk diligence, customer activation and assistance for faster productionization, as well as upselling and cross-selling of Plaid products
• You will have a front row seat to all the latest industry developments
• Over time, with the skills and experience you develop and hone on this team, you can become an influential voice in defining where AI <> Fintech will be heading longer term
• Work with other AI engineers, software engineers and machine learning engineers to architect, design and implement GenAI-powered products and features
• Collaborate across functions to understand user needs, propose and implement AI-powered solutions where they’re expected to have the highest impact
• Design and execute rapid experiments to push the boundaries on potential business impact from emerging AI capabilities, with a focus on minimal viable testing approaches
• Balance creative exploration of possibilities with rigorous evaluation of technical feasibility, product potential and business impact