FalconX
FalconX is an AI voice agent-based hotel bid booking system that automates the entire process of hotel search, price comparison, negotiation and ordering through parallel dialing, structured quote parsing and dual-track pricing.
Timeline (last 90 days)
FalconX is a project that replaces manual phone calls with AI voice agents, handing the entire hotel booking chain of search—price comparison—negotiation—ordering to a real-time orchestration system.
FalconX won second prize at the TechTO Hackathon.
FalconX's design principles are distilled into three measurable goals: speed, price, and distance.
FalconX uses FastAPI as its orchestration core, parsing natural-language booking requests into structured fields (destination, check-in/check-out dates, number of guests, email address).
FalconX's voice interaction layer is handled by ElevenLabs Conversational AI, with Twilio providing the telephone line connection.
FalconX's backend orchestrator handles quotes and confirmations through structured tool calls rather than relying on plain-text transcription, thereby distinguishing conversational responses from reliable booking decisions.
FalconX employs a parallel dialing strategy, contacting multiple high-potential candidate hotels simultaneously and continuously updating the best quote as quotes come back.
FalconX's pricing layer integrates SerpApi for hotel search data and LiteAPI for bookable prices, with the two serving as cross-references.
FalconX has prepared a hotel database focused on Ontario, California and surrounding communities, storing hotel details and phone numbers in advance to reduce the latency of discovering candidates and obtaining quotes.
FalconX's web interface is built with Next.js, displaying hotel locations, candidate prices, call progress, call transcripts, and the currently leading quote in real time.
1 more timeline events
All Facts (11)
FalconX's infrastructure is deployed on Google Cloud Run.
50%UnverifiedFalconX is a project that replaces manual phone calls with AI voice agents, handing the entire hotel booking chain of search—price comparison—negotiation—ordering to a real-time orchestration system.
50%UnverifiedFalconX's design principles are distilled into three measurable goals: speed, price, and distance.
50%UnverifiedFalconX uses FastAPI as its orchestration core, parsing natural-language booking requests into structured fields (destination, check-in/check-out dates, number of guests, email address).
50%UnverifiedFalconX's voice interaction layer is handled by ElevenLabs Conversational AI, with Twilio providing the telephone line connection.
50%UnverifiedFalconX's backend orchestrator handles quotes and confirmations through structured tool calls rather than relying on plain-text transcription, thereby distinguishing conversational responses from reliable booking decisions.
50%UnverifiedFalconX employs a parallel dialing strategy, contacting multiple high-potential candidate hotels simultaneously and continuously updating the best quote as quotes come back.
50%UnverifiedFalconX's pricing layer integrates SerpApi for hotel search data and LiteAPI for bookable prices, with the two serving as cross-references.
50%UnverifiedFalconX has prepared a hotel database focused on Ontario, California and surrounding communities, storing hotel details and phone numbers in advance to reduce the latency of discovering candidates and obtaining quotes.
50%UnverifiedFalconX's web interface is built with Next.js, displaying hotel locations, candidate prices, call progress, call transcripts, and the currently leading quote in real time.
50%UnverifiedFalconX won second prize at the TechTO Hackathon.
50%