I am building an end-to-end AI layer for my car-rental and fleet-management platform and need someone who can design, train, and deploy three cooperating agents. The first agent oversees booking management—quoting availability, handling modifications, flagging clashes and pushing confirmed reservations into our existing database. A second agent focuses on live fleet monitoring: it watches real-time GPS feeds, cross-checks vehicle maintenance records and instantly alerts staff or customers when a car is due for service or strays outside a geo-fence. The third handles customer service, chatting with renters 24/7 through a web-based interface to answer policy questions, extend rentals or raise support tickets. All three agents will work from the same data lake that already stores customer-booking history, maintenance logs and the live location stream we ingest from on-board trackers. You’ll need to design the data pipelines, craft the conversational logic (NLU/NLP), set confidence thresholds and deploy to our Kubernetes cluster so the bots scale with demand. Deliverables • Conversational flows and intents for the chat interface • Backend services (Python / Node preferred) that fetch and write to booking, maintenance and GPS tables via our REST API • Deployment scripts (Docker & Helm) plus basic monitoring dashboards • A short hand-over guide and video walkthrough showing the agents running in our staging environment I’ll provide schema docs, sample datasets and API keys the moment we start. If you’ve built multi-agent systems with Dialogflow, Rasa, LangChain, or direct OpenAI calls—and can demonstrate production uptime—let’s talk.