Job Type: Full time

Project Role : AI Infrastructure ArchitectProject Role Description : Architect and build custom Artificial Intelligence (AI) infrastructure/hardware solutions. Optimize AI infrastructure/hardware performance, power consumption, cost and scalability of computational stack. Advise on AI infrastructure technology and vendor evaluation, selection and full stack integration.Must have skills : Agent DevelopmentGood to have skills : NAMinimum 7.5 year(s) of experience is requiredEducational Qualification : 15 years full time educationLevel Senior LevelExperience 5 to 8 YearsWork Mode Remote or HybridEmployment Full TimeKey ResponsibilitiesBuild multi agent orchestration systems usingVertex AIAgent BuilderLangGraphAutoGenDesign scalable and resilient pipelines integratingLLM reasoningTool callingMemory managementHuman in the loop flowsLead the design and implementation of RAG systems usingVertex AI SearchMatching EngineEnterprise vector storesDefine and enforce GCP infrastructure standards using a security first approach acrossGKECloud RunEventarcApigeeDrive MLOps practices includingModel evaluationAgent observabilityTracingLoggingContinuous improvement loopsPartner with product,



data science, and platform teams to translate business requirements into agentic solutionsConduct in depth code reviewsConduct architecture reviewsMentor developers with 3 to 5 years of experienceContribute toInternal developer platformsSDKsReusable agent componentsRequired Skills and Qualifications5 to 8 years of software or platform engineering experience2 or more years focused on AI ML or agentic systemsDeep expertise in GCP services includingVertex AIGeminiVertex AI PipelinesFeature StoreGKEPub SubProven experience building production grade LLM powered applications withTool useMulti turn reasoningAgent memoryProficiency in PythonStrong understanding ofAsync patternsConcurrencyDistributed systemsHands on experience with agentic frameworks such asADKLangChainLangGraphCrewAIExperience withVector databasesEmbedding pipelinesHybrid search architecturesStrong command of GCP security includingIAMWorkload IdentityVPC Service ControlsSecret ManagerTrack record of deliveringHigh availability servicesLow latency servicesSLO SLA ownership

📌 Ai Infrastructure Architect (Pune)
🏢 Accenture
📍 Pune