About the Role
We are looking for a Senior Android Engineer to build and scale large, multi-module Android applications used by a large and demanding user base. You will work at the intersection of mobile engineering and applied AI, owning build architecture, UI performance, app stability, and the integration of LLM driven features into production.
This role needs someone who thinks in systems but ships in sprints. You will make architectural decisions that affect several feature teams, while staying close enough to the code to diagnose an ANR at 2am or cut build times by half.
Key Responsibilities
Architecture and Build Systems
- Design and maintain scalable multi-module Android architectures that let multiple feature teams work independently without breaking each other.
- Own the Gradle build setup, including convention plugins, version catalogs, build logic modularisation, and dependency management.
- Optimise build performance through configuration caching, build caching, parallel execution, and module graph tuning.
- Define and enforce module boundaries, API contracts, and dependency rules across the codebase.
UI Development
- Build high performance, accessible UIs with Jetpack Compose.
- Optimise recomposition, state handling, and rendering performance for complex screens.
- Contribute to and maintain a shared design system or component library.
Stability, Performance, and Monitoring
- Implement and maintain Firebase Crashlytics for crash reporting, alerting, and triage.
- Diagnose and resolve memory leaks, ANRs, jank, and startup performance issues with Android Studio Profiler, LeakCanary, Perfetto, and similar tools.
- Set up performance baselines and monitor regressions across releases.
Dependency Injection and Testability
- Apply clean dependency injection patterns using Hilt or Dagger across modules.
- Write testable code and maintain unit, integration, and UI test coverage.
AI and LLM Integration
- Integrate LLM APIs (Claude, OpenAI, or similar) into production mobile features.
- Build and consume MCP (Model Context Protocol) servers and tools within production workflows.
- Handle token management, context window strategy, streaming responses, and cost control on mobile.
- Implement agentic workflows and multi-agent orchestration where the product requires it.
- Design AI error recovery: timeouts, retries, fallbacks, hallucination handling, and graceful degradation when the model or network fails.
- Balance AI capability against real mobile constraints such as latency, battery, bandwidth, and offline behaviour.
Collaboration
- Work closely with product, design, backend, and AI/ML teams to deliver features end to end.
- Mentor engineers, review code, and raise the technical bar across feature teams.
- Document architectural decisions and share knowledge internally.
Mandatory Requirements
- 6+ years of native Android development in Kotlin.
- Proven experience with large scale, multi-module Android applications.
- Strong Gradle expertise, including build optimisation and custom build logic.
- Production experience with Jetpack Compose.
- Hands-on experience with Hilt or Dagger.
- Firebase Crashlytics implementation and crash analysis experience.
- Proven ability to diagnose memory issues, ANRs, and performance bottlenecks.
- Hands-on production experience integrating LLM APIs.
- Working experience with MCP frameworks or tool servers.
- Solid understanding of token management, context handling, agentic workflows, and AI error recovery.
Technical Expertise (at least 2 to 3 of the following)
- Kotlin Coroutines and Flow
- Jetpack DataStore and Room
- OkHttp and Protocol Buffers
- WorkManager
Preferred Qualifications
- Active GitHub profile with multiple Android projects demonstrating clean architecture and scalability.
- Open source contributions to Android or AI tooling.
- Technical writing: blogs, conference talks, or internal engineering documentation.
- Experience shipping AI powered features to real users at scale.
- Familiarity with on-device ML (ML Kit, TensorFlow Lite, Gemini Nano) is a plus.
- Experience in banking, fintech, or other regulated environments is a plus.
What Success Looks Like in the First 6 Months
- Measurable reduction in build times across the project.
- Reduced crash rate and ANR rate on key user journeys.
- At least one AI powered feature shipped to production with proper monitoring and fallback handling.
- Clear, documented module structure that feature teams can build on without friction.