I need an end-to-end AI platform built for mass adoption that can automate repetitive tasks at scale. The core goal is to let non-technical users launch, monitor, and refine automations through a clean web interface while the underlying machine-learning models continually improve from usage data. Scope of work The first milestone is a proof of concept that ingests a CSV or REST feed, detects recurring patterns, and then auto-generates an executable workflow or bot to handle those steps. Think of a self-learning alternative to traditional RPA tools—cloud-native, containerised, and ready for thousands of concurrent users. Key expectations • Modular micro-service architecture on AWS, GCP, or Azure • Python or Node.js core with well-documented REST APIs • RBAC dashboard showing workflow history, success metrics, and manual override controls • Robust logging, error handling, and CI/CD via GitHub Actions (or similar) Acceptance criteria 1. A sample dataset triggers automated workflow creation within 5 minutes. 2. Automation handles ≥1,000 parallel calls with sub-200 ms latency. 3. Dashboard clearly reports run history, success rates, and supports on-the-fly edits or retraining. While the initial release is industry-agnostic, the architecture must remain flexible so future modules (e.g., HL7 healthcare parsers or fintech KYC routines) can slot in without major refactoring. Deliverables include source code, Docker files, IaC scripts, and setup documentation packaged for smooth hand-off.