Principal AI Automation Engineer/Lead (Fully Remote)
Please mention DailyRemote when applying
AI Summary
You will own the end-to-end architecture of the AI automation pipeline, focusing on perception, AI orchestration, and dynamic execution. This includes developing mechanisms for data extraction, architecting prompt strategies, and building a dynamic execution engine.
*This role is fully remote within TAIWAN ONLY.
The Mission
We are spearheading a strategic architectural pivot from Web to Windows Desktop Automation. We are looking for a visionary Lead Engineer to architect and build a proprietary "Text-to-Code" generation engine from the ground up.
In this role, you will bridge the gap between complex Windows OS structures and Large Language Models (LLMs) to create a self-healing, intelligent automation agent that perceives, reasons, and executes tasks autonomously.
Key Responsibilities
You will own the end-to-end architecture of the AI automation pipeline, focusing on three core pillars:
1. Perception & Data Synthesis
Develop robust mechanisms to traverse and extract data from complex, multi-layered Windows UI trees.
Design intelligent "Signal-to-Noise" filtering algorithms to sanitize UI data, ensuring critical information is preserved while staying within LLM context window limits.
Define standardized data schemas to transform raw system states into AI-ready inputs.
2. AI Orchestration & Prompt Engineering
Architect sophisticated prompt strategies to guide LLMs in generating high-quality, executable automation code.
Implement defensive coding patterns within the AI logic to handle desktop-specific challenges such as latency, asynchronous loading, and UI state fluctuations.
Continuously optimize the interaction between system data and model reasoning to minimize "hallucinations" and maximize script accuracy.
3. Dynamic Execution & Runtime
Build a dynamic execution engine capable of compiling and running AI-generated code in real-time within memory.
Develop a comprehensive Test Runner that manages execution lifecycles, exception handling, and automated reporting (e.g., state capturing and error logging).
Ensure the scalability of the engine to support diverse desktop application architectures.
Must-Have Requirements
- Expertise in C# / .NET: Advanced proficiency in C# with experience in low-level system interactions, reflection, or dynamic code compilation.
- Windows UI Automation Depth: Proven experience with Windows UI automation frameworks (UIA3 or similar) and a deep understanding of the Windows OS accessibility layer.
- Generative AI & LLM Integration: Practical experience integrating LLMs into production software workflows, including advanced prompt engineering and context management.
- Architectural Leadership: Demonstrated ability to design "Zero-to-One" technical pipelines and define engineering standards for complex, novel systems.
Nice-to-Have Skills (Preferred Qualifications)
- VDI & Remote Environments: Experience navigating automation constraints within - Virtual Desktop Infrastructure (e.g., resolution locking, headless execution).
- Advanced Algorithm Design: Experience with recursive tree traversal and data optimization algorithms.
- Legacy & Modern Desktop Frameworks: Familiarity with the underlying structures of WinForms, WPF, or UWP applications.
- Testing Infrastructure: Deep understanding of unit testing frameworks (e.g., NUnit, xUnit) and their application in automated CI/CD pipelines.
Why This Role?
High Autonomy: As the principal owner, you define the technical roadmap, the JSON schemas, and the AI prompt strategy.
Solve the "Unsolvable": Desktop automation is notoriously brittle. You will use Generative AI to solve the classic "Element Not Found" problem by building a truly intelligent, adaptive system.
Innovation at Scale: You aren't just maintaining code; you are building the engine that defines the future of AI-driven quality engineering.
The Mission
We are spearheading a strategic architectural pivot from Web to Windows Desktop Automation. We are looking for a visionary Lead Engineer to architect and build a proprietary "Text-to-Code" generation engine from the ground up.
In this role, you will bridge the gap between complex Windows OS structures and Large Language Models (LLMs) to create a self-healing, intelligent automation agent that perceives, reasons, and executes tasks autonomously.
Key Responsibilities
You will own the end-to-end architecture of the AI automation pipeline, focusing on three core pillars:
1. Perception & Data Synthesis
Develop robust mechanisms to traverse and extract data from complex, multi-layered Windows UI trees.
Design intelligent "Signal-to-Noise" filtering algorithms to sanitize UI data, ensuring critical information is preserved while staying within LLM context window limits.
Define standardized data schemas to transform raw system states into AI-ready inputs.
2. AI Orchestration & Prompt Engineering
Architect sophisticated prompt strategies to guide LLMs in generating high-quality, executable automation code.
Implement defensive coding patterns within the AI logic to handle desktop-specific challenges such as latency, asynchronous loading, and UI state fluctuations.
Continuously optimize the interaction between system data and model reasoning to minimize "hallucinations" and maximize script accuracy.
3. Dynamic Execution & Runtime
Build a dynamic execution engine capable of compiling and running AI-generated code in real-time within memory.
Develop a comprehensive Test Runner that manages execution lifecycles, exception handling, and automated reporting (e.g., state capturing and error logging).
Ensure the scalability of the engine to support diverse desktop application architectures.
Must-Have Requirements
- Expertise in C# / .NET: Advanced proficiency in C# with experience in low-level system interactions, reflection, or dynamic code compilation.
- Windows UI Automation Depth: Proven experience with Windows UI automation frameworks (UIA3 or similar) and a deep understanding of the Windows OS accessibility layer.
- Generative AI & LLM Integration: Practical experience integrating LLMs into production software workflows, including advanced prompt engineering and context management.
- Architectural Leadership: Demonstrated ability to design "Zero-to-One" technical pipelines and define engineering standards for complex, novel systems.
Nice-to-Have Skills (Preferred Qualifications)
- VDI & Remote Environments: Experience navigating automation constraints within - Virtual Desktop Infrastructure (e.g., resolution locking, headless execution).
- Advanced Algorithm Design: Experience with recursive tree traversal and data optimization algorithms.
- Legacy & Modern Desktop Frameworks: Familiarity with the underlying structures of WinForms, WPF, or UWP applications.
- Testing Infrastructure: Deep understanding of unit testing frameworks (e.g., NUnit, xUnit) and their application in automated CI/CD pipelines.
Why This Role?
High Autonomy: As the principal owner, you define the technical roadmap, the JSON schemas, and the AI prompt strategy.
Solve the "Unsolvable": Desktop automation is notoriously brittle. You will use Generative AI to solve the classic "Element Not Found" problem by building a truly intelligent, adaptive system.
Innovation at Scale: You aren't just maintaining code; you are building the engine that defines the future of AI-driven quality engineering.
Automatically Apply to the Best Remote Jobs
Stop the endless job search. Our AI finds and applies to the best jobs for you.
