Voice AI startup · co-founded · 2024 – 2025
Voice AI & Multi-Agent Customer Automation Platform
Co-Founder & CEO
Challenge
Enterprises needed a voice-AI-first platform for inbound and outbound customer interactions—not single prompt chains, but collaborative agents that reason through complex queries, plug into CRM and ops systems, and deploy on voice and messaging without rebuilding workflows per channel. Workflow automation with LLMs breaks when one agent handles every step; coordination, hybrid retrieval, and failure recovery need explicit multi-agent design. Vertical products on the same runtime had to handle mixed structured and unstructured queries (e.g. realty brokers waiting days for answers spanning listings and project docs).
Approach
- Built a voice-AI-first platform unifying touchpoints under agentic workflows: train on business data, configure actions, deploy on voice, WhatsApp, and similar channels from one agent definition.
- Designed collaborative multi-agent architecture with explicit coordination strategies—specialized agents reason with each other instead of monolithic single-flow bots; applied MARL-informed handoff patterns for robustness under real conversation load.
- Engineered hybrid decision-routing between structured (SQL) and unstructured (RAG) sources so agents pick the right retrieval path per query intent.
- Shipped enterprise integration layer (150+ connectors) so agents act proactively inside existing systems with minimal human oversight.
- Deployed vertical products on the shared runtime—including a realty communication layer combining Supabase listings and Qdrant document embeddings, cutting broker query turnaround from two days to minutes.
Tech stack
LLM agents · Voice AI / telephony · WhatsApp & messaging · Multi-agent orchestration · Hybrid RAG + SQL routing · Supabase · Qdrant · CRM integrations · Workflow automation · Python · TypeScript
Outcomes
- End-to-end platform: train → configure actions → deploy → evaluate agents across voice and chat
- Production multi-agent workflow runtime, not demo-level prompt chains
- 150+ enterprise integrations for sales and support automation
- Broker query turnaround reduced from 2 days to minutes on realty vertical deployment
- Reusable hybrid retrieval and coordination patterns for additional verticals
Context
B2B customer interaction automation and vertical SaaS on shared multi-agent infrastructure. Customer and employer names anonymized.