Client work · Voice AI · April–October 2025
Connecting voice agents to a working call-center platform.
Full-stack work across conversational workflows, LiveKit voice agents, tool calling, knowledge retrieval, and SIP telephony.
- Product
- Voho AI
- My role
- Full Stack Engineer
Delivered contribution
Contributed agent workflows and integrations to Voho’s white-label call-center platform for sales, recruiting, reception, and support.
- 01
Inbound or outbound call
- 02
Conversation + knowledge retrieval
- 03
Tool / function call
- 04
Connected business workflow
The product problem
A call-center agent needs to do more than generate a spoken answer. It needs a phone connection, access to relevant knowledge, and a way to act on the business workflow behind the conversation. The web product, backend, and voice runtime all have a part in that experience.
Voho’s white-label platform brought those surfaces together for different customer workflows. My role covered the full stack and cloud layer, rather than a standalone voice demo.
What I worked on
I built conversational workflows and LiveKit-powered voice agents, including tool and function calling, knowledge-base retrieval, and SIP integrations for inbound and outbound calling. I also worked across the web product, backend services, and cloud infrastructure.
Those responsibilities connected the caller-facing agent to the systems around it. Retrieval supplied business knowledge; tool calls connected conversation to actions; SIP connected the workflow to telephony. The product surface and backend supported how those capabilities were used within the platform.
Why the integration work matters
A convincing conversation is one part of a voice product. The rest is whether the agent can use the information and actions the business actually depends on. This engagement required working across those boundaries, rather than treating speech, application code, and deployment as separate deliverables.
For a buyer evaluating similar work, the useful question is which parts of the workflow need custom engineering: the call flow, the business-system connection, the application around it, or all three.
Delivered contribution
My contribution supported white-label deployments for sales, recruiting, reception, and support workflows. It was work within an existing team and platform, not sole ownership of the entire product.
The linked founder endorsement supports the scope and quality of my engineering work. This case study describes my 2025 engagement; Voho’s current product may have evolved since then. No private customer data or unverified revenue, latency, or conversion figures are included.
Explore the work
Building a voice workflow?
Tell me about your call flow and the systems it needs to connect to. We can start with the part that needs engineering.
Discuss a similar project ↗