Staff Software Engineer - Digital Experience Platform

About Nu

Nu serves more than 140 million customers, guided by a mission to fight complexity and empower people. The company has been leading an industry transformation through innovative products and human-centered services.

Proprietary technology and data at scale power Nu’s digital platform, built to promote financial access, advancement, and transparency. Its business model thrives on customer love and lower costs, feeding a flywheel of growth and profitability. Visit our Institutional Page

About the role

The Digital Experience Platform (DXP) is Nubank’s AI-first platform ecosystem designed to radically compress our product development workflow from concept to production.

Today, building “Magic App”-quality experiences in our Backend-Driven Content (BDC) stack takes ~49 working days and ~24 handoff-heavy steps — DXP exists to close this Creation Gap and Workflow Gap by making AI-native tooling the default way we design, build, and ship experiences.

We are building four core products: DXP Studio (an AI-powered creation environment that turns Figma designs into production-ready BDC code), DXP Dictionary (the curated knowledge base that makes AI agents “Nu-aware”), DXP Spec Protocol (our vendor-agnostic abstraction layer), and DXP Workflow (an orchestration platform for hybrid human + AI agent teams across the product lifecycle).

As a Staff Software Engineer – DXP , you will be the technical leader and multiplier for this initiative: shaping the architecture of an AI-native platform, driving critical decisions across the Figma-to-BDC pipeline and agent orchestration, and partnering closely with Product, Design, and Engineering leadership to prove value and scale the platform company-wide.

We’re looking for someone genuinely addicted to AI — who obsessively explores new models, agent frameworks, and AI-powered tools, and can turn that passion into leverage for thousands of Nubank engineers.

You’ll be responsible for

Architect the DXP Studio pipeline: Design and lead the end-to-end architecture for the Figma → DXP Spec → BDC code generation flow, ensuring vendor-agnostic abstractions (DXP Spec Protocol), NuDS component fidelity, and production-ready output quality.

Lead AI agent integration & orchestration: Define how AI agents (codegen, QA, design-mapping, etc.) are composed, executed, and supervised inside DXP; drive decisions on LLM provider abstraction, context-assembly strategies, MCP tool surfaces, and execution environments.

Own the DXP Dictionary and knowledge architecture: Architect the knowledge base that turns generic AI into Nubank-aware agents — how NuDS specs, tokens, canonical patterns and domain knowledge are ingested, structured, and consumed by generation and validation pipelines.

Drive cross-platform integration: Lead the technical integration model with NuDS (component catalog & tokens), BDC/Core (patterns & services), HPP (widget / homepage infrastructure), and App Excellence (observability & quality metrics), defining clear contracts that position DXP as an orchestrator on top of existing platforms.

Establish quality-at-generation-time: Design automated quality enforcement for generated experiences — NuDS compliance, token correctness, accessibility rules, structural integrity, golden tests — so governance happens by default at generation time, not via manual post-hoc review.

Set technical strategy and roadmap: Contribute to the 6–18 month technical strategy for DXP, translating strategic bets into executable plans, identifying architectural risks and trade-offs between POCs and long-term scalability.

Multiply engineering excellence: Mentor engineers within DXP and partner teams (NuDS, BDC, Mobile Platform), lead design reviews and RFCs, and raise the bar for AI-augmented development practices across Nubank.

We are looking for a person who has

Must Have

Genuine obsession with AI tools and workflows — deep, hands-on, daily experience with LLMs, AI agents and modern AI-powered dev tools (e.g., Claude Code, Cursor, Copilot, custom agents, MCP servers); you are the person who has tried new tools before others even ask about them.

Strong software architecture skills , with experience designing multi-component platform systems that integrate AI/LLM capabilities into production workflows.

Deep understanding of design-to-code pipelines, design systems or component-driven UI development , ideally bridging design tooling (e.g., Figma) and code generation.

Practical experience with AI/LLM integration in software products : prompt engineering, agent orchestration, context management, RAG architectures and/or MCP tool development.

Proven ability to lead cross-functional technical initiatives involving multiple teams and stakeholders, driving alignment and delivery in ambiguous environments.