Eranga et al. 2025 characterize production agentic workflows as dynamic pipelines of multiple specialized agents using different LLMs, tool-augmented capabilities, and orchestration logic, and they offer a structured engineering lifecycle plus nine core best practices—including tool-first design over MCP and single-responsibility agents—for building them.
E4 describes agentic workflows as dynamic pipelines integrating multiple specialized agents with different LLMs, tool-augmented capabilities, and orchestration logic, and E5 states the structured engineering lifecycle and nine core best practices including tool-first design over MCP and single-tool/single-responsibility agents.
Written by Kimi K3 via Ollama Cloud · checked by GLM-5.3 via Ollama Cloud · 1 Oct, 04:35
Source chain
AEvery quote below was checked, without a model, to appear verbatim in its source.
- 01
“agentic workflows integrate multiple specialized agents with different Large Language Models(LLMs), tool-augmented capabilities, orchestration logic, and external system interactions”
Full-text passage · p. 1 · evidence E4
A Practical Guide for Designing, Developing, and Deploying Production-Grade Agentic AI Workflows Eranga Bandaraa, Ross Gore a, Peter Foytika, Sachin Shettya, Ravi Mukkamalaa, Abdul Rahman b, Xueping Liang c, Safdar H. Bouk a, Amin Hassg, Sachini Rajapakse f, Ng Wee Keong d, Kasun De Zoysa e, Aruna Withanageh, Nilaan Loganathan h aOld Dominion University, Norfolk, VA, USA bDeloitte & Touche LLP, USA cFlorida International University, USA dNanyang Technological University, Singapore eUniversity of Colombo, Sri Lanka fIcicleLabs.AI gAnaletIQ, VA, USA hEffectz.AI Abstract Agentic AI marks a major shift in how autonomous systems reason, plan, and execute multi-step tasks. Unlike traditional single model prompting, agentic workflows integrate multiple specialized agents with different Large Language Models(LLMs), tool-augmented capabilities, orchestration logic, and external system interactions to form dynamic pipelines capable of autonomous decision-making and action.
Passage read from arxiv.org, which may be a preprint rather than the published version.
- 02
“We introduce a structured engineering lifecycle encompassing workflow decomposition, multi-agent design patterns, Model Context Protocol(MCP), and tool integration, deterministic orchestration”
Full-text passage · p. 2 · evidence E5
ing, developing, and deploying production-quality agentic AI systems. We introduce a structured engineering lifecycle encompassing workflow decomposition, multi-agent design patterns, Model Context Protocol(MCP), and tool integration, deterministic orchestration, Responsible-AI considerations, and environment-aware deployment strategies. We then present nine core best practices for engineering production-grade agentic AI workflows, including tool-first design over MCP, pure-function invocation, single-tool and single-responsibility agents, externalized prompt management, ResponsibleAI-aligned model-consortium design, clean separation between workflow logic and MCP servers, containerized deployment for scalable operations, and adherence to the Keep it Simple, Stupid (KISS) principle to maintain simplicity and robustness. To demonstrate these principles in practice, we present a comprehensive case study: a multimodal news-analysis and media-generation workflow. By combining architectural guidance, operational patterns, and practical implementation insights, this paper offers a foundational reference to build robust, extensible, and production-ready agentic AI workflows.
Passage read from arxiv.org, which may be a preprint rather than the published version.
- 03
“nine core best practices for engineering production-grade agentic AI workflows, including tool-first design over MCP, pure-function invocation, single-tool and single-responsibility agents”
Full-text passage · p. 2 · evidence E5
ing, developing, and deploying production-quality agentic AI systems. We introduce a structured engineering lifecycle encompassing workflow decomposition, multi-agent design patterns, Model Context Protocol(MCP), and tool integration, deterministic orchestration, Responsible-AI considerations, and environment-aware deployment strategies. We then present nine core best practices for engineering production-grade agentic AI workflows, including tool-first design over MCP, pure-function invocation, single-tool and single-responsibility agents, externalized prompt management, ResponsibleAI-aligned model-consortium design, clean separation between workflow logic and MCP servers, containerized deployment for scalable operations, and adherence to the Keep it Simple, Stupid (KISS) principle to maintain simplicity and robustness. To demonstrate these principles in practice, we present a comprehensive case study: a multimodal news-analysis and media-generation workflow. By combining architectural guidance, operational patterns, and practical implementation insights, this paper offers a foundational reference to build robust, extensible, and production-ready agentic AI workflows.
Passage read from arxiv.org, which may be a preprint rather than the published version.