Lead AI / Agentic Platform Architect -EST Time Zone

Reflections Info Systems Pvt Ltd; Thiruvananthapuram; Remote; INR 11,407,000 - 17,110,000; Full time

Job summary

Reflections Info Systems Pvt Ltd seeks an experienced Lead AI / Agentic Platform Architect to design and engineer scalable, governance‑driven AI platforms powered by LLMs and agentic technologies. You will architect agent orchestration, memory, and tool integrations while ensuring safety and security.

You will evaluate architectures, establish standards, and provide hands‑on technical leadership across engineering teams in EST time zone operations.

Qualifications

• 5+ years in software architecture, AI engineering, or platform engineering with hands-on LLM/agentic AI experience.
• Strong grasp of agent orchestration frameworks, runtimes, multi-agent workflows.
• Hands-on with LLM integrations, prompt engineering, function calling, and tool orchestration.
• Experience with memory, state management, and context handling for AI systems.
• Knowledge of vector databases, embeddings, semantic search, and retrieval mechanisms.
• Experience building guardrails, evaluation frameworks, and observability for AI platforms.

Responsibilities

• Design scalable, enterprise-grade agentic AI platforms and solutions.
• Define orchestration patterns, runtimes, workflows, and multi-agent communication.
• Architect prompt engineering, tool calling, connectors, and model integration strategies.
• Design memory/state management and context handling mechanisms.
• Evaluate and implement RAG architectures and vector DB integrations.

Skills

LLM integration

Agent orchestration

Multi-agent systems

Prompt engineering

Memory management

RAG architectures

Vector databases

AI security governance

Communication & collaboration

Tools

LangChain

LangGraph

Semantic Kernel

AutoGen

CrewAI

Azure OpenAI

AWS Bedrock

Google Vertex AI

Pinecone

Weaviate

Qdrant

Azure AI Search

Job description

We are looking for an experienced Lead AI / Agentic Platform Architect with strong hands‑on expertise in designing, architecting, and engineering LLM-powered and agentic AI systems.The ideal candidate should have practical experience building and orchestrating AI agents, integrating LLMs, designing agent runtimes, implementing tool integrations, and establishing scalable, secure, and governed AI platforms.This role requires someone who can evaluate existing agentic AI architectures, understand platform dependencies, define technical standards, establish evaluation frameworks, and identify gaps in safety, security, and AI governance.The candidate should be equally comfortable with architecture‑level decisions and hands‑on technical implementation.

Work Schedule:

Must be flexible and willing to work in the US Eastern Standard Time (EST) time zone.

Responsibilities include:

Agentic AI Architecture & Platform Engineering

• Design and architect scalable, enterprise‑grade agentic AI platforms and solutions.
• Define agent orchestration patterns, runtime architecture, workflows, and multi‑agent communication mechanisms.
• Architect prompt engineering frameworks, tool calling, connectors, and model integration strategies.
• Design memory management, state persistence, and context management mechanisms.
• Evaluate and implement RAG architectures and vector database integrations wherever applicable.

LLM & AI Integration

• Design and integrate LLM‑based applications using leading commercial and open‑source models.
• Establish model integration patterns, routing, fallback mechanisms, and model selection strategies.
• Define reusable components and frameworks for agent development and deployment.
• Optimize model performance, latency, scalability, and token consumption.

AI Evaluation, Monitoring & Observability

• Define evaluation frameworks to measure agent accuracy, reliability, consistency, and performance.
• Establish mechanisms for testing prompts, tools, workflows, and end‑to‑end agent behaviour.
• Implement telemetry, monitoring, tracing, and observability for agentic systems.
• Define mechanisms to identify hallucinations, failures, and performance bottlenecks.

AI Safety, Security & Governance

• Identify and address security and governance gaps in agentic AI architectures.
• Establish guardrails for responsible AI implementation, including access controls and data protection.
• Define policies for tool permissions, human‑in‑the‑loop controls, and agent execution boundaries.
• Ensure appropriate compliance, auditability, and governance standards are incorporated into the architecture.

Technical Leadership & Architecture Assessment

• Assess existing AI and agentic platform architectures and map technical dependencies.
• Identify architecture gaps, integration challenges, scalability risks, and opportunities for improvement.
• Recommend target‑state architecture and implementation roadmaps.
• Collaborate with engineering, product, and business stakeholders to translate requirements into technical solutions.
• Provide technical guidance and architectural direction to engineering teams.

Primary Skills :

• 5+ years of experience in software architecture, AI engineering, platform engineering, or related technical roles, including recent hands‑on experience building LLM‑powered and agentic AI systems.
• Strong understanding of agent orchestration frameworks, agent runtimes, multi‑agent systems, and workflow automation.
• Hands‑on experience with LLM integrations, prompt engineering, function calling, and tool orchestration.
• Experience designing memory, state management, context handling, and RAG-based architectures.
• Knowledge of vector databases, embeddings, semantic search, and retrieval mechanisms.
• Experience implementing AI guardrails, evaluation frameworks, and observability solutions.
• Strong understanding of AI security, governance, data privacy, and responsible AI principles.
• Ability to assess complex architectures, identify dependencies, and recommend scalable technical solutions.
• Strong communication and stakeholder management skills.

Secondary Skills :

• Experience with frameworks and platforms such as LangChain, LangGraph, Semantic Kernel, AutoGen, CrewAI, or equivalent.
• Exposure to cloud‑based AI ecosystems such as Azure OpenAI, AWS Bedrock, or Google Vertex AI.
• Experience with vector databases such as Pinecone, Weaviate, Qdrant, or Azure AI Search.
• Familiarity with agent evaluation tools, tracing, and monitoring frameworks.
• Experience designing enterprise AI platforms with security, scalability, and cost optimization considerations.

Role:

Lead AI / Agentic Platform Architect -EST Time Zone

Close Date:

06-11-2026