Agentic AI Platform Architect & Product Lead

Position - Agentic AI Platform Architect & Product Lead
Budget – 20-50LPA per annum (high level)
Experience – 10-15 years.
Location – Hyderabad
Primary Skill – Communication should be top notch.
Employment Type - Full-Time / Contractor

Experience
10 years of overall software engineering experience with 3 years of hands-on experience in AI/ML LLM application development Generative AI platform architecture solution architecture and technical product role where you help take a technical platform accelerator or IP to market as a commercial offering.

Role Overview
We are looking for an experienced Agentic AI Platform Architect & Product Lead to support the development and evolution of the Fluid AIQ Platform and lead AI-powered application development for upcoming client projects.
The ideal candidate should have strong hands-on experience building production-ready applications using LLMs AI Agents RAG tool/function calling workflow orchestration and modern AI frameworks.
This role requires someone who can work beyond simple chatbot implementations and help design agentic applications capable of reasoning planning using tools interacting with enterprise systems and operating within defined guardrails.

The candidate will work closely with the architecture and development teams contribute to the Fluid AIQ Platform evaluate emerging AI technologies build POCs and take technical and product ownership of Fluid platform components and client AI programs. Equally important this is a product and commercialization role: the candidate will shape how Fluid is brought to market and how its components are packaged so that each commercial module can be deployed and sold on its own and then expanded as a client’s needs grow while reporting platform health adoption and delivery metrics to leadership.

Key Responsibilities
1. AI / Agentic AI Architecture
• Design and develop scalable architectures for Agentic AI applications and lead and mentor architects and engineers in setting engineering standards for agentic delivery.
• Identify appropriate use cases for AI agents versus traditional application workflows.
• Design multi-agent and single-agent architectures based on business requirements.
• Define agent orchestration tool usage memory state management and workflow patterns and capture them as reusable reference architectures client onboarding playbooks implementation guides and sales-enablement material.
• Evaluate AI technologies and recommend the right architecture models frameworks and infrastructure applying a replaceable-execution principle that keeps platform-owned context governance economics and evaluation logic independent of any single model agent framework coding agent or cloud.
2. Agentic Application Development
Develop AI applications involving: AI Agents and Multi-Agent Systems Agent orchestration Tool/function calling Workflow automation Human-in-the-Loop (HITL) RAG Long/short-term agent memory Planning and reasoning workflows MCP Server / MCP Client integrations Enterprise API and system integrations AI-powered decision support Autonomous and semi-autonomous workflows.
3. LLM & Generative AI
• Work with leading LLM providers such as: OpenAI Anthropic / Claude Azure OpenAI Google Gemini Open-source models Ollama / local models.
• Design model-selection strategies based on cost performance latency privacy and use case.
• Implement structured outputs function calling prompt engineering context management and model fallbacks.
• Evaluate and benchmark different models for specific business use cases.
4. AI Frameworks & Technologies
Hands-on experience with frameworks and technologies such as: LangChain LangGraph LlamaIndex AI Agents / Agent SDKs MCP Vector databases RAG frameworks AI evaluation and observability tools.
The candidate should be able to evaluate frameworks rather than being dependent on a single technology.
5. AI Safety Guardrails & Reliability
• Implement AI guardrails and validation mechanisms.
• Design controls for hallucination incorrect tool usage unauthorised actions and unsafe outputs.
• Implement human approval workflows where required.
• Design reliable agent workflows with error handling retries fallback mechanisms and observability supported by the release management environments SLAs / SLOs and support model needed to run Fluid as a product rather than a series of projects.
• Consider security privacy compliance and responsible AI requirements.

Required Technical Skills
Core AI / GenAI
Strong understanding of: Generative AI LLMs AI Agents / Agentic AI RAG Prompt Engineering Function / Tool Calling Embeddings Vector Search Context Management AI Evaluation.
Agentic AI
Strong practical understanding of: Agent architecture Agent orchestration LangGraph or equivalent Tool calling Multi-agent systems Agent state and memory Human-in-the-loop workflows Autonomous / semi-autonomous workflows Agent guardrails Agent observability.
Programming
Strong programming skills in at least one of: / TypeScript Python.
Additional experience with: REST APIs Microservices Authentication / authorization Async processing Event-driven architectures Database integration.
AI Frameworks
Hands-on experience with one or more: LangChain LangGraph LlamaIndex OpenAI SDK Anthropic SDK Claude Agents SDK Other Agent SDKs.
Models
Experience integrating multiple LLM providers preferably: OpenAI Anthropic / Claude Azure OpenAI Gemini Ollama / open-source LLMs.
Data & RAG
Experience with: PostgreSQL Vector databases pgvector Pinecone / Weaviate / Qdrant / similar Embedding models Document processing Semantic search Hybrid search Knowledge graphs & ontology design (Neo4j or similar RDF / OWL GraphRAG) Multi-tenant data architecture.
Cloud
Strong experience with at least one cloud platform: Microsoft Azure AWS Google Cloud.

Nice-to-Have Skills
Experience with: MCP (Model Context Protocol) Claude Code Codex / AI coding tools GitHub Copilot Cursor Ollama Open-source LLMs Multimodal AI Voice AI AI workflow automation AI observability / evaluation platforms Fine-tuning Model evaluation AI security Prompt injection protection PII / data protection Enterprise SaaS integrations CRM / ERP integrations AI FinOps / FinOps Foundation practices Cloud marketplace listings (Azure / AWS Marketplace) Prior product management role Productizing consulting IP into licensed software.
RPA / Automation Platforms
Experience replacing or complementing traditional automation/RPA solutions with AI-agent-based approaches is particularly valuable.

Required Skills:

Agentic AI Architecture LLM & Generative AI RAG (Retrieval-Augmented Generation) Tool / Function Calling Prompt Engineering & Context Management AI Evaluation & Benchmarking AI Frameworks AI Safety Guardrails & Reliability Programming Data & Vector Databases AWS Azure GCP Enterprise Integration Platform Productization Commercialization & Go-to-Market Technical Leadership & Mentoring