Senior AI / Agentic Architect

Comviva; Bengaluru; On-site; INR 4,000,000 - 6,000,000; Full time

Job summary

Comviva in Bengaluru seeks a senior Architect to design enterprise-grade Agentic Orchestration Platforms. You will craft scalable multi-agent architectures, manage agent lifecycles, memory, and planning, and implement autonomous tool-using agents and RAG systems.

Responsibilities include setting standards for prompt engineering, AI evaluation, governance, and security. The role covers cloud-native platforms on AWS/Azure/GCP and Kubernetes-based runtimes, with an emphasis on IDP practices and GPU

Qualifications

• Strong experience in Agentic AI systems and multi-agent orchestration.
• Hands-on expertise with LLM orchestration frameworks and RAG architectures.
• Proficiency with LangChain, LangGraph, Semantic Kernel, CrewAI, AutoGen and related tooling.
• Commercial cloud/platform skills across AWS, Azure, GCP and modern containerized runtimes.

Responsibilities

• Define and architect enterprise-grade Agentic Orchestration Platforms.
• Design scalable architectures for multi-agent orchestration, agent lifecycle management, memory and context handling, goal decomposition and planning.
• Design and implement autonomous and tool-using agents, RAG systems, multi-agent collaboration patterns, AI copilots and decision intelligence systems.
• Establish architecture standards for prompt engineering, context injection, memory frameworks, AI evaluation and observability, and responsible AI controls.
• Architect cloud-native AI platforms on AWS, Azure, or GCP.
• Design scalable runtime environments using Kubernetes, container orchestration, service mesh, API gateways, and event-driven frameworks.
• Drive best practices for Internal Developer Platforms (IDP), GitOps & IaC, AI workload scalability, GPU-aware scheduling, and performance optimization.
• Define enterprise-wide AI governance frameworks for responsible AI usage, explainability and auditability, data privacy and compliance, security guardrails, and model access control mechanisms.

Skills

Agentic AI systems

Multi-agent orchestration

LLM orchestration

LangChain/LangGraph

Kubernetes & Docker

AWS/Azure/GCP

Python/FastAPI

API design

Vector databases

AI governance

Job description

• Define and architect enterprise-grade Agentic Orchestration Platforms
• Design scalable architectures for multi-agent orchestration, agent lifecycle management, agent memory and context handling, goal decomposition and planning, autonomous workflow execution, human-in-the-loop governance, and agent communication frameworks
• Design and implement autonomous and tool-using agents, Retrieval-Augmented Generation (RAG) systems, multi-agent collaboration patterns, AI copilots and decision intelligence systems, and AI Ops agents
• Establish architecture standards for prompt engineering, context injection, agent memory frameworks, AI evaluation and observability, hallucination mitigation, and responsible AI controls
• Architect cloud-native AI platforms on AWS, Azure, or GCP
• Design scalable runtime environments using Kubernetes, container orchestration, service mesh, API gateways, and event-driven frameworks
• Drive best practices for Internal Developer Platforms (IDP), GitOps & Infrastructure as Code, AI workload scalability, GPU-aware scheduling, and performance optimization
• Define enterprise-wide AI governance frameworks for responsible AI usage, explainability and auditability, data privacy and compliance, security guardrails, and model access control mechanisms

Key Responsibilities

• Define and architect enterprise-grade Agentic Orchestration Platforms
• Design scalable architectures for multi-agent orchestration, agent lifecycle management, agent memory and context handling, goal decomposition and planning, autonomous workflow execution, human-in-the-loop governance, and agent communication frameworks
• Design and implement autonomous and tool-using agents, Retrieval-Augmented Generation (RAG) systems, multi-agent collaboration patterns, AI copilots and decision intelligence systems, and AI Ops agents
• Establish architecture standards for prompt engineering, context injection, agent memory frameworks, AI evaluation and observability, hallucination mitigation, and responsible AI controls
• Architect cloud-native AI platforms on AWS, Azure, or GCP
• Design scalable runtime environments using Kubernetes, container orchestration, service mesh, API gateways, and event-driven frameworks
• Drive best practices for Internal Developer Platforms (IDP), GitOps & Infrastructure as Code, AI workload scalability, GPU-aware scheduling, and performance optimization
• Define enterprise-wide AI governance frameworks for responsible AI usage, explainability and auditability, data privacy and compliance, security guardrails, and model access control mechanisms

Required Skills

• Strong experience in Agentic AI systems, multi-agent orchestration, AI workflow automation, context-aware AI architectures, generative AI platforms, and autonomous systems design
• Hands-on expertise in LLM orchestration frameworks, RAG architectures, vector databases, AI memory frameworks, and AI evaluation pipelines
• Experience with LangChain, LangGraph, Semantic Kernel, CrewAI, AutoGen, Model Context Protocol (MCP), AI Gateway frameworks, agent lifecycle frameworks, and AI observability platforms
• Strong expertise in AWS / Azure / GCP, Kubernetes & Docker, microservices architecture, event-driven systems (Kafka, streaming platforms), API-first design, platform engineering, GitOps, and Terraform
• Strong coding experience in Python, FastAPI, Node.js / TypeScript (preferred), distributed systems engineering, and API integration frameworks

Key Competencies

• Strategic thinking with hands-on architecture capability
• Strong problem-solving and system design skills
• Ability to work across cross-functional teams
• Deep understanding of scalable enterprise platforms
• Focus on innovation, governance, and performanc