Staff Software Engineer - Digital Experience Platform

# **About Nu** Nu is the leading digital bank in Latin America, serving 140 million customers across Brazil, Mexico, and Colombia. The company has been leading an industry transformation by leveraging data and proprietary technology to develop innovative products and services. Guided by its mission to fight complexity and empower people, Nu caters to customers’ complete financial journey, promoting financial access and advancement with responsible lending and transparency. The company is powered by an efficient and scalable business model that combines low cost to serve with growing returns. Nu’s impact has been recognized in multiple awards, including Time 100 Most Influential Companies, Fast Company’s Most Innovative Companies, and Forbes World’s Best Banks. Visit our [Institutional Page](https://www.nu.com/2026-en) ### 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. - Excellent **communication skills**, translating complex architectural decisions into clear narratives for Product, Design and senior leadership. - **Product-minded mindset**, balancing technical excellence with pragmatic delivery — comfortable proving value via POCs before scaling platform capabilities. - Experience with **Nubank’s stack or similar patterns**: BDC (Backend-Driven Content), and broader mobile/frontend ecosystems. Will be a Plus - Experience with **Flutter** at the intersection of design systems and code generation. - Hands-on experience building or integrating **AI coding agents** (e.g., Claude Code, Devin, Cursor) into team workflows. - Familiarity with **workflow orchestration platforms**, agent farms or multi-agent systems. - Understanding of **analytical architectures**, experimentation platforms, feature flags and progressive rollout strategies for generated experiences. - Prior experience productizing **internal developer tools or platforms** that serve large engineering orgs. - Contributions to **open-source AI developer tools** or active presence in the AI-for-developers community. ### Location for this opportunity This role can be based in one of the following hubs: - **Toronto, Canada** - **São Paulo, Brazil** - **Rio de Janeiro, Brazil** - **Campinas, Brazil** - **Belo Horizonte, Brazil** ### Work Model for this Role - **Hybrid 2–3 times/week**: Our hybrid work model brings us to the office at least twice a week, on strategic days designed to maximize team connection and collaboration. For more details, visit ### Our Benefits – Global Roles For global, multi-location roles like this, our benefits package is competitive in each country and generally includes: - Competitive compensation and the opportunity of **earning equity** at Nu - Health / Medical Insurance - Life Insurance (and AD&D where applicable) - Extended maternity and paternity leaves - **Nucleo** – our learning platform of courses - **NuLanguage** – our language learning program - **NuCare** – our mental health and wellness assistance program - Paid time off and local holidays - Retirement / savings plans (e.g., 401(k), HSA/FSA in the US; country-specific plans elsewhere) - Work-from-home allowance - Relocation assistance package, if applicable Details for each location will be shared during the process, aligned with local regulations and practices. ### Join us If you want to define how an AI-native bank designs and ships product experiences — turning agentic workflows, design systems and code generation into everyday tools for hundreds of teams — we’d love to hear from you. Our recruitment process may involve the use of artificial intelligence–enabled tools, such as automated interview transcription and analysis, to support the evaluation process. Artificial intelligence is not used to make final hiring decisions; all decisions are made by human reviewers.