AI Agents · Real Estate
AI agents that answer real estate calls. Phone, Chat, WhatsApp.
Rebuilt Nesti's v2 platform with the founding team - multi-channel AI agents and real-time call analytics for real estate teams, shipped on weekly sprints.
What we built
- Voice AI
- Production phone agent
- Multi-channel
- Phone + Chat + WhatsApp on one AI stack
- Analytics
- Call trends, caller types, AI insights
Engagement
The challenge
Real estate teams miss calls. A lot of them. Leads come in at all hours — evenings, weekends, during showings — and every missed call is a potential deal lost. Nesti had a working v1 product but needed a complete platform rebuild to support multiple communication channels and deliver the kind of AI-powered call handling that could actually replace a human receptionist.
How we approached it
We joined the founding team for the v2 rebuild, working in weekly sprints with fast iteration cycles. The goal was clear: build AI agents that could handle inbound calls on phone, chat, and WhatsApp — all from a single unified platform. We started with the voice AI pipeline, getting a production-quality phone agent live first, then expanded to chat and WhatsApp. Each channel had its own quirks — latency requirements for voice, rich media for WhatsApp, real-time typing indicators for chat — but they all needed to share the same AI brain.
The technical work
The voice AI system needed to handle real-time speech-to-text, process the intent through an LLM, generate a contextual response, and push it back through text-to-speech — all within the conversational latency window that makes phone calls feel natural. We built the orchestration layer that manages this pipeline, handles interruptions, and gracefully falls back when the AI is uncertain. The analytics dashboard gives property managers visibility into call volumes, caller types, common questions, and AI performance — turning every interaction into actionable data.
Where it is today
Nesti handles calls across three channels from a single platform. Real estate teams use it to ensure no lead goes unanswered, with AI agents that know their listings, can schedule viewings, and qualify prospects. The partnership is now over two years running, with continuous improvements to the AI models and expansion into new real estate verticals.
The people behind this project
Client
Will Fletcher Co-founded Nesti and led the product strategy. Defined the multi-channel vision and worked directly with real estate teams to validate the AI agent experience against real-world call handling needs.
Drove the go-to-market and customer onboarding. Worked closely with early real estate partners to gather feedback and shape the product around actual agent workflows.
Handled the business development side — identifying target verticals within real estate and building the partnerships pipeline that fed product direction.
Zeroic
Prashant Architected the v2 platform rebuild and the real-time voice AI pipeline. Designed the orchestration layer that handles speech-to-text, LLM inference, and text-to-speech within conversational latency windows.
Anish Built the multi-channel integration layer — connecting phone, chat, and WhatsApp to the shared AI brain. Owned the analytics dashboard that gives property managers visibility into call patterns and AI performance.
Building a voice AI or multi-channel product?
We built Nesti's real-time voice agents from scratch — phone, chat, and WhatsApp on one stack. If you're tackling conversational AI, we should talk.
Book a call