Senior Core Flutter Engineer (AI Tooling / Harness)

We are looking for a Senior Core Flutter Engineer to join a high-impact engineering initiative focused on AI tooling and harness capabilities for the mobile platform. In this context, a Flutter AI Harness is the control and context layer that transforms a generic AI into a Flutter specialist that understands the project rules, architecture constraints, and engineering conventions of the codebase it operates on. This role is not primarily about shipping consumer-facing AI features inside the app. It is about building the internal foundations, workflows, and agentic tooling that allow engineers to automate migrations, reason about codebases, and operate safely at scale across Flutter and BDC-heavy environments.

As a Senior Engineer, you will work across Flutter and Dart tooling, structured specs, LLM integrations, static analysis, plugin or agent surfaces, and evaluation workflows that help engineering teams operate faster and more safely. You are expected to bring 5+ years of overall software engineering experience, with at least 4+ years of strong hands-on Flutter and Dart experience, plus meaningful exposure to AI-assisted developer workflows or platform tooling. The ideal candidate combines mobile platform depth with strong judgment around guardrails, observability, and human-in-the-loop automation in complex engineering environments.

Requirements

• Flutter + Dart Platform Depth: Strong hands-on experience with Flutter and Dart at platform level, including libraries, tooling, package structure, analyzer constraints, and engineering workflows beyond feature-only mobile development.

• AI Harness Framing: Clear understanding that the harness is not just an LLM wrapper, but the control and context layer that makes an AI behave like a senior Flutter specialist inside a specific project and its rules.

• Agentic AI / LLM Tooling: Practical experience building AI-assisted tooling or agentic workflows that use LLMs in a loop, including structured outputs, tool use, retries, and context management for engineering tasks.

• LLM Provider Integration: Hands-on experience integrating providers such as OpenAI, Anthropic, Gemini, or gateway-based equivalents, with secure authentication patterns and operational awareness of latency, cost, and failure modes.

• PII and LLM Data Handling: Practical understanding of what content cannot safely reach third-party LLM providers, prompt hygiene, and how to design agents that fail safe when sensitive engineering or business context is in scope.

• Spec-Driven Workflows: Experience using machine-readable specifications, structured task definitions, or similar patterns to drive planning, execution, and validation for automated engineering workflows.

• Static Analysis of Dart Code: Ability to work with the Dart analyzer, AST visitors, linting patterns, or equivalent code-analysis techniques to support migrations, refactors, or engineering automation.

• Plugin or Tooling Surfaces: Experience building IDE, CLI, plugin, or internal tooling surfaces that expose developer workflows cleanly and can be integrated into existing engineering environments.

• MCP and Tool Exposure: Familiarity with MCP-style tool interfaces or equivalent patterns for exposing tools safely to agents, including considerations around permissions, data boundaries, and user trust.

• Security-Aware Agent Design: Strong judgment around guardrails, human review boundaries, sensitive data handling, and the distinction between what an agent may automate directly versus what must remain gated.

• Observability for AI Systems: Experience instrumenting LLM-based workflows with logs, metrics, traces, auditability, token or cost awareness, and debugging signals that make AI tooling supportable in production-like environments.

• Evaluation and Regression Testing: Commitment to validating AI-assisted workflows through repeatable tests, evaluation datasets, human review loops, or other mechanisms that reduce regressions and unsafe automation.

• Mobile Platform Fluency: Comfort working in complex Flutter mobile ecosystems that include shared tooling, platform constraints, and integration with broader mobile architecture rather than isolated app features.

• Developer Experience Mindset: Ability to build tooling that reduces engineering toil, improves migration speed, and helps teams move safely through repetitive or high-friction platform work.

• English C1: Ability to communicate technical trade-offs clearly, write design notes, and collaborate effectively in a remote, multicultural environment.

EXTRAS (Nice to have)

• Fleet-Scale Migration Experience: Experience operating batch migrations, multi-package refactors, or scorecard-style tooling across many packages or repositories rather than only one app at a time.

• Navigation or BDC Migration Context: Familiarity with navigation migration patterns, server-driven UI environments, or BDC-style ecosystems can be valuable when adapting AI tooling to mobile platform workflows.

• Evaluation Engineering: Exposure to eval datasets, LLM-as-judge approaches, human-in-the-loop review flows, or other quality systems for AI-assisted development tooling.

• Riverpod / go_router / Mobile Architecture Patterns: Additional familiarity with the platform patterns most likely to appear in migration and automation tasks is useful when the tooling must reason about real code at scale.

• IDE and CLI Extensions: Experience shipping internal extensions, plugins, or command-line tooling for engineering teams is valuable when the harness needs to live inside existing workflows.

Quick Skills Reference

• Languages: Dart, TypeScript or Python, JSON / YAML-style structured specs

• Mobile: Flutter, Dart Tooling, Mobile Platform Architecture, Shared Libraries

• AI / LLM: OpenAI, Anthropic, Gemini, Structured Outputs, Tool Use, Evaluations

• Tooling: Dart Analyzer, AST Workflows, IDE / CLI Plugins, MCP-Style Integrations

Highlights