Browse categories
Explore
Fiverr Pro
English
$
USD
I will build enterprise ai agents, multi agent orchestrators, n8n workflows
Oleksii Y
About this gig
Multi-Agent AI Orchestrator for Jira & QA Automation
Stop wasting engineering hours on routine ticket triaging, requirements analysis, and test design. I will build a modular, enterprise-grade AI Multi-Agent System using n8n, Ollama, and Qdrant to automate your SDLC.
How the Architecture Works:
The system uses a micro-workflow design where a central "QA Orchestrator" acts as a router, monitors Jira, and dynamically triggers specialized sub-agents:
- Requirements Agent: Evaluates user stories, flags gaps, and generates clarification questions.
- Test Case Agent: Automatically drafts detailed test cases (ID, Priority, Steps, Expected Results tables).
- Checklist Agent: Generates traceable functional and non-functional verification checklists.
- Bug Enricher Agent: Optimizes raw bug reports for developer usability.
️ Production Features:
Context-Aware RAG: Connected to Qdrant Vector Store to retrieve your project-specific standards.
Micro-Workflows: Sub-agents run via n8n toolWorkflow nodes for easy updates and debugging.
100% Local AI: Compatible with Ollama (Qwen, Llama) for corporate data security and GDPR.
Get to know Oleksii Y
Oleksii Y
AI Native Quality Assurance Engineer
- FromRomania
- Member sinceApr 2026
- Avg. response time3 hours
Languages
Ukrainian, Russian, English
Transforming Quality Assurance through AI-Driven Strategy & Deep Audit.
With 10+ years of experience in the QA trenches, I don't just "find bugs"—I build resilient quality ecosystems. I specialize in auditing complex software architectures and implementing AI-native workflows that accelerate testing cycles by up to 300%.
FAQ
Is my company’s data (Jira tickets, source docs) safe? Can we run this locally?
Absolutely. The architecture is fully compatible with local open-source LLMs running via Ollama (like Qwen, Llama, or Mistral) and a local Qdrant instance. No data leaves your infrastructure, which ensures 100% GDPR compliance and corporate data security.
Why do you build this using multiple sub-workflows instead of a single n8n flow?
A single monolithic flow breaks easily and is impossible to maintain at scale. By using n8n's toolWorkflow nodes, I build a micro-service architecture. The main QA Orchestrator handles routing, while dedicated agents (Test Case Creator, Bug Report Enricher) run in completely isolated workflows.
How do the AI agents know about our specific project rules and standards?
he system leverages RAG (Retrieval-Augmented Generation) connected to a Qdrant Vector Store. Before any sub-agent (like the QA or BA agent) generates an output, it automatically queries the vector database using vector embeddings to pull your project's specific templates and terminologies.
What do I need to provide to get this system up and running?
You will need an active n8n instance (Cloud or Self-Hosted via Docker), access to your Jira API/Webhooks, Slack/Teams (if human-in-the-loop is needed), and API credentials for your chosen LLM provider or local Ollama host. If you don't have n8n deployed yet, you can select my "Self-Hosted n8n" extra
Can we modify the instructions or templates that the AI agents use?
Yes. All system instructions, quality checklists, and formatting rules are stored clearly within the agent prompts or inside the Qdrant database. I will show you how to tweak these parameters so you can easily adjust the outputs to fit changing project requirements.
