Sanskar Kharya

An AI agent that remembers the caller

Nothing frustrates a customer more than repeating themselves on every call.

A lot of teams already manage their leads and customer data in Google Sheets. Exporting that data, uploading a CSV, and then figuring out how to get call results back into the same workflow adds unnecessary steps.

So we built a direct Google Sheets integration into Vaani AI Campaigns. @Vaani Developer Platform

Here s what you can do:

• Connect your Google account: Pick a spreadsheet and the specific tab you want to use.

• Map your columns: Map phone numbers, names, and agent variables like city or plan type to Vaani s contact fields.

• Control what gets written back: Choose columns for call summaries, status, disposition, duration, and extracted information.

• Launch your campaign: Vaani picks up contacts from the sheet and starts calling.

• Keep it in sync: Once calls are processed, their results are written back to the corresponding rows. Add new rows later and they can be picked up on the next sync without importing everything again.

One thing that s easy to overlook when building voice agents is what happens when a provider goes down or a model starts returning errors mid conversation.

In a normal API workflow, a failed request might mean retrying or showing an error. In a live voice call, it s different. The caller is waiting, and a failure in one component shouldn t bring down the entire conversation.

At Vaani AI @Vaani Developer Platform ,we ve been working on a multi-layer failover mechanism to handle this:

• Independent fallback chains: Speech recognition, LLM and TTS each have their own backup providers.

• Error-based switching: If the primary provider returns an error, the request moves to the next configured provider.

• Safety-net fallback: Even when an agent is misconfigured, there s a fallback model available.

• Configurable routing: Teams can choose primary and backup models, use their own API keys, and control whether fallback is enabled.