Job Title

Description and Requirements Key ResponsibilitiesWindows Computer Use & AutomationDesign and implementagent actionsthat interact with Windows OS, applications, and system servicesBuild reliablecomputer use primitivessuch as application control, system configuration, file operations, and window managementIntegrate with Windows APIs, services, and background tasksEnsure agents function correctly across user sessions, power states, and device conditionsLLM Integration & Agent IntelligenceIntegrateLLMs into agent workflowsfor task planning, decision-making, and intent interpretationImplement prompt templates, tool-calling patterns, and structured outputsHandle fallbacks between LLM-based reasoning and deterministic logicOptimizeLLM usage for latency, cost, and reliabilityAgent Execution & ReliabilityImplement execution logic including retries, rollback, and error handlingOptimizesystem performance, power usage, and responsivenessSupport long-running and event-driven agent workflowsSecurity, Privacy & TrustRespect Windows security boundaries and permission modelsImplement safeguards to ensure transparent and user-controlled automationContribute to logging, auditing, and explainability of agent actionsCollaborationPartner closely with Agent Platform, AI, UX, and Android teamsParticipate in architecture reviews and cross-platform design discussionsHelp define best practices for AI-driven computer use automationRequired Qualifications4+ years of professional software engineering experienceStrong experience buildingWindows desktop or system-level softwareProficiencyinC++ and/or C# (.NET)Hands-on experience withWindows Services, background execution, or system APIsAt least 2 years of experience integrating and using Large Language Models (LLMs)in production or near-production systems, including:Calling LLM APIs or frameworksIntegrating LLMs into application workflowsHandling structured outputs, error cases, and latency considerationsStrong understanding ofmultithreading, concurrency, and system debuggingPreferred QualificationsExperience withagent-based automation or computer use systemsAdvanced LLM integration experience, including:Tool / function callingPrompt engineering and evaluationHybrid rule-based + LLM architecturesKnowledge ofWin32, WinRT, UWP, WMI, or PowerShellFamiliarity with:Power management and resource constraintsEnterprise or OEM Windows environmentsExposure toAI safety, guardrails, or explainability techniques