AI Engineer

Eidiko Systems Integrators; Hyderabad; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

Eidiko Systems Integrators is seeking an AI Agent Architect to design, build, and deploy intelligent agents across low-code platforms. You will craft prompts, ensure safe behavior, and forge workflows that connect to enterprise systems via REST/SOAP APIs and A2A protocols.

The role requires building orchestration patterns, coding in Python/JS, and designing RAG pipelines for knowledge-grounded agents. You’ll collaborate with stakeholders and maintain robust documentation.

Qualifications

• Design, build, test, and deploy AI agents on low-code / no-code platforms like Microsoft Copilot Studio, N8N, Make, Helix Studio, Oracle AI Studio, and other emerging agentic platforms.
• Craft high-quality system prompts, user prompts, and instruction sets that guide agent behaviour including chain-of-thought reasoning, tool-calling directives, guardrails, and fallback logic.
• Architect agentic workflows encompassing multi-step reasoning, tool use, memory management, context grounding, and human-in-the-loop escalation patterns.
• Integrate AI agents with enterprise systems (CRM, ERP, ITSM, HRMS, core banking, etc.) via REST/SOAP APIs, MCP servers, A2A protocol, Power Automate connectors, webhooks, and custom plugins.
• Build and manage multi-agent orchestration patterns including delegation, sub-agent routing, conditional branching, and cross-platform interoperability using A2A and similar protocols.
• Develop custom scripts (Python, JavaScript/TypeScript) for data transformation, API middleware, custom MCP servers, or evaluation harnesses.
• Design and implement Retrieval-Augmented Generation (RAG) pipelines for knowledge-grounded agents, including document ingestion, chunking strategies, embedding, and vector store integration.
• Create structured evaluation frameworks for agents — including golden test sets, Exact Match, LLM-as-a-Judge scoring, prompt regression testing, and automated quality gates.
• Collaborate with business stakeholders to identify automation opportunities, define agent scope, and translate requirements into actionable agent specifications.
• Maintain comprehensive documentation of agent architectures, prompt libraries, integration patterns, and deployment runbooks.
• Stay current with the rapidly evolving agentic AI landscape — new platforms, protocols (MCP, A2A, ACP), frameworks (LangChain, LangGraph, AutoGen, Semantic Kernel, CrewAI), and best practices.
• Support governance and compliance requirements by implementing appropriate guardrails, content filters, audit trails, and role-based access controls within agent deployments.

Responsibilities

• Design, build, test, and deploy AI agents on low-code / no-code platforms.
• Craft prompts, directives, guardrails, and fallback logic for agent behavior.
• Architect multi-step agent workflows with memory management and escalation.
• Integrate agents with enterprise systems via REST/SOAP APIs, MCP, and connectors.
• Create multi-agent orchestration with delegation, routing, and cross-platform interoperability.
• Develop Python/JavaScript scripts for data transformation and middleware tasks.
• Design RAG pipelines for knowledge grounding including document ingestion and embeddings.
• Build evaluation frameworks with golden tests and automated gates.
• Collaborate with stakeholders to convert requirements into agent specs.
• Document architectures, prompt libraries, integration patterns, and runbooks.
• Keep up to date with agentic AI platforms, protocols, and frameworks.
• Implement governance measures like guardrails and audit trails in deployments.

Skills

AI agent design

Low-code/no-code

Python

JavaScript/TypeScript

REST/SOAP APIs

MCP servers

RAG pipelines

LLM evaluation

Tools

LangChain

LangGraph

AutoGen

Semantic Kernel

CrewAI

Weights & Biases

Application Insights

Datadog

Job description

Role & responsibilities

• Design, build, test, and deploy AI agents on multiple low-code / no-code platforms

like Microsoft Copilot Studio, N8N, Make, Helix Studio, Oracle AI Studio, and other

emerging agentic platforms.

• Craft high-quality system prompts, user prompts, and instruction sets that guide

agent behaviour including chain-of-thought reasoning, tool-calling directives,

guardrails, and fallback logic.

• Architect agentic workflows encompassing multi-step reasoning, tool use, memory

management, context grounding, and human-in-the-loop escalation patterns.

• Integrate AI agents with enterprise systems (CRM, ERP, ITSM, HRMS, core

banking, etc.) via REST/SOAP APIs, MCP servers, A2A protocol, Power Automate

connectors, webhooks, and custom plugins.

• Build and manage multi-agent orchestration patterns including delegation, sub-

agent routing, conditional branching, and cross-platform agent interoperability using

A2A and similar protocols.

• Develop custom scripts (Python, JavaScript/TypeScript) for scenarios where low-

code capabilities are insufficient — e.g., data transformation, API middleware,

custom MCP servers, or evaluation harnesses.

• Design and implement Retrieval-Augmented Generation (RAG) pipelines for

knowledge-grounded agents, including document ingestion, chunking strategies,

embedding, and vector store integration.

• Create structured evaluation frameworks for agents — including golden test sets,

Exact Match, LLM-as-a-Judge scoring, prompt regression testing, and automated

quality gates.

• Collaborate with business stakeholders to identify automation opportunities, define

agent scope, and translate requirements into actionable agent specifications.

• Maintain comprehensive documentation of agent architectures, prompt libraries,

integration patterns, and deployment runbooks.

• Stay current with the rapidly evolving agentic AI landscape — new platforms,

protocols (MCP, A2A, ACP), frameworks (LangChain, LangGraph, AutoGen,

Semantic Kernel, CrewAI), and best practices.

• Support governance and compliance requirements by implementing appropriate

guardrails, content filters, audit trails, and role-based access controls within agent

deployments.

Preferred candidate profile

• Microsoft certifications: AI-102 (Azure AI Engineer), PL-200 (Power Platform Functional Consultant), or the upcoming AB-620 (Copilot Studio AI Agent Builder).
• UiPath certifications: UiPath Certified Automation Developer or Agentic Automation Developer Associate.
• Experience with agentic frameworks such as LangChain, LangGraph, AutoGen, Semantic Kernel, or CrewAI.
• Familiarity with evaluation and observability tools: LangSmith, Weights & Biases, Application Insights, Datadog.
• Experience building custom MCP servers or implementing A2A agent discovery and task delegation.
• Knowledge of responsible AI principles, content filtering, PII handling, and AI governance frameworks.
• Exposure to RPA (Robotic Process Automation) tools and their integration with agentic AI workflows.
• Multi-language LLM prompting experience (English + Arabic is a plus for a Dubai-based role).