Agentic AI Architect

About Position:
We are seeking a highly experienced Agentic AI Architect with 10-15 years of industry experience to lead the design, architecture, and implementation of next-generation AI-driven solutions. The ideal candidate will possess deep expertise in Agentic AI frameworks, large language models (LLMs), autonomous AI systems, and cloud-native architectures. This role involves collaborating with business and technology stakeholders to define AI strategies, architect scalable intelligent agents, and deliver enterprise-grade AI solutions leveraging Vertex AI and modern cloud platforms.
Role: Agentic AI Architect
Location: Hyderabad & Bengaluru
Experience: 10 to 15 Years
Job Type: Full Time Employment
What You'll Do:
Study the current business process, including users, decisions, handoffs, systems, information sources, exceptions, delays, controls and failure points.
Separate problems that need AI reasoning from those better solved through process correction, rules, search, analytics, workflow automation or conventional software.
Define the target process and show exactly where an AI assistant or agent participates, where deterministic controls apply and where a person must review or approve.
Convert business goals into measurable outcomes such as reduced handling time, improved first-time accuracy, fewer manual handoffs, faster information retrieval, lower rework or better control compliance.
Establish the current KPI baseline, target improvement, measurement method, data source, reporting frequency and business owner before claiming value.
Explain expected capabilities, limitations, uncertainty, failure modes, dependencies and residual risks in language understood by business and technology stakeholders.
Design end-to-end architectures covering models, RAG, knowledge sources, agent orchestration, tools, APIs, state, memory, user channels, identity, permissions and human escalation.
Define clear boundaries for agent autonomy, including which actions are read-only, which require confirmation and which must never be delegated to an agent.
Design for secure data access, tenant separation, privacy, audit evidence, prompt-injection resistance, least-privilege tools and controlled handling of sensitive information.
Define the evaluation strategy, including representative business scenarios, test datasets, acceptance thresholds, tool-action checks, safety tests and regression gates.
Define production requirements for response time, availability, throughput, cost, logging, tracing, quality monitoring, incident handling, rollback and model or prompt changes.
Select appropriate models, retrieval approaches and frameworks based on quality, security, latency, cost, portability and operating constraints rather than vendor preference.
Demonstrate deep proficiency in at least one AI development stack while adapting designs to enterprise-developed platforms, standards and custom frameworks.
Build or guide technical prototypes to test the highest-risk assumptions before committing to full implementation.
Review solution designs, code, prompts, tool contracts, data flows, test evidence and production-readiness material produced by engineering teams.
Create architecture diagrams, decision records, capability boundaries, risk assessments, cost models, delivery roadmaps and operating-model documentation.
Provide technical direction to AI engineers, QA engineers, data engineers, application engineers, security specialists and operations teams.
Track whether deployed solutions achieve the agreed KPIs and recommend tuning, redesign or retirement when the expected value is not being realized.
Expertise You'll Bring:
10+ years in software, data, cloud or solution architecture, including substantial hands-on experience with production AI or LLM-based systems.
Strong understanding of business-process analysis, value-stream mapping, decision points, control requirements and measurable outcome design.
Deep technical knowledge of LLMs, prompting, structured generation, embeddings, vector and hybrid search, reranking, RAG, tool use, agent orchestration, state and memory.
Demonstrable expertise in at least one production AI stack, such as Vertex AI and Gemini with ADK, LangChain/LangGraph, Semantic Kernel, LlamaIndex, or an equivalent platform.
Ability to review Python code, APIs, data pipelines, retrieval logic, agent workflows, automated tests and deployment designs at an engineering level.
Experience designing integrations with enterprise applications, APIs, data platforms, identity systems and approval workflows.
Strong understanding of AI evaluation, including answer quality, source support, task completion, tool-call correctness, safety, latency and cost.
Experience designing cloud-native, containerized or serverless solutions with CI/CD, secrets, IAM, observability and rollback controls.
Strong knowledge of responsible AI, privacy, secure AI architecture and controls for systems capable of taking actions.
Ability to create simple value models that distinguish projected benefit from measured benefit and avoid unsupported ROI claims.
Strong workshop facilitation, consulting, written communication and architecture-presentation skills.
Ability to challenge unsuitable AI use cases constructively and recommend a simpler solution where appropriate.
Production architecture experience with GCP services such as Vertex AI, Gemini, Vertex AI Search, BigQuery, Cloud Storage, Cloud Run, GKE, Pub/Sub, Cloud Logging, Cloud Monitoring and Cloud Trace.
Experience with knowledge graphs, document intelligence, multimodal systems, MCP or equivalent tool-integration protocols.
Experience designing AI solutions in regulated environments with strict audit, privacy, residency and approval requirements.
Familiarity with model and agent evaluation frameworks, experiment tracking and production AI monitoring.
Relevant cloud, machine-learning, generative AI or architecture certification.
Current-process and target-process views showing human, system and agent responsibilities.
AI suitability assessment explaining why an agent-based approach is or is not justified.
Capability statement documenting what the solution will do, may do with approval and will not do.
KPI definition containing baseline, target, measurement source, reporting cadence and accountable business owner.
End-to-end architecture and data-flow diagrams, including trust and permission boundaries.
Evaluation, security, observability, cost and production-support strategies.
Architecture decisions, identified risks, assumptions, dependencies and phased delivery roadmap.
Benefits:
Competitive salary and benefits package
Culture focused on talent development with quarterly growth opportunities and company-sponsored higher education and certifications
Opportunity to work with cutting-edge technologies
Employee engagement initiatives such as project parties, flexible work hours, and Long Service awards
Annual health check-ups
Insurance coverage: group term life, personal accident, and Mediclaim hospitalization for self, spouse, two children, and parents
Values-Driven, People-Centric & Inclusive Work Environment:
Persistent is dedicated to fostering diversity and inclusion in the workplace. We invite applications from all qualified individuals, including those with disabilities, and regardless of gender or gender preference. We welcome diverse candidates from all backgrounds.
We support hybrid work and flexible hours to fit diverse lifestyles.
Our office is accessibility-friendly, with ergonomic setups and assistive technologies to support employees with physical disabilities.
If you are a person with disabilities and have specific requirements, please inform us during the application process or at any time during your employment
Let's unleash your full potential at Persistent - persistent.com/careers
"Persistent is an Equal Opportunity Employer and prohibits discrimination and harassment of any kind."

Work arrangement
No