Share Your Requirements
Teleshop Ltd.
LANCERProject Description
The client engaged Oodles to build an advanced LLM-based orchestration system designed to automate end-to-end sourcing workflows. The solution leverages a local LLM orchestrator to manage multiple modular automation units, each responsible for a specific task within the sourcing pipeline. The goal was to create a scalable, intelligent system capable of executing, monitoring, and optimizing complex workflows through a centralized dashboard.
Scope Of Work
The project focused on developing a backend-driven orchestration system along with a dashboard interface to monitor and control workflow execution. It included designing modular ATOMs for task-specific automation, integrating a local LLM for orchestration, and customizing an existing dashboard for tracking workflow progress. The scope also covered enabling manual intervention, real-time status visibility, and seamless coordination between different automation units.
Our Solution
Oodles delivered a robust and modular AI orchestration system by combining LLM capabilities with structured automation workflows.
Key implementations included:
- LLM-Based Orchestration: Implemented a local LLM (Ollama) to intelligently control and sequence multiple automation units.
- Modular ATOM Architecture: Designed independent ATOMs, each handling a specific sourcing function for scalability and flexibility.
- Workflow Automation: Enabled end-to-end automation of sourcing processes through coordinated execution of ATOMs.
- Dashboard Integration: Customized a Gmail-like dashboard interface to monitor workflow progress and status in real time.
- Real-Time Tracking & Visibility: Implemented step-by-step tracking of sourcing flows for better transparency and control.