Skip to content
Notis

Find the customer problems hiding in call transcripts

Review the needs customers actually described, with enough context to understand each one.

Trigger

Webhook received

Notis starts this workflow when an external tool or custom backend sends an HTTP request.

Action

Create Chat Completion

Tool to create a chat completion using DeepSeek models (deepseek-chat or deepseek-reasoner). Use when you need to generate AI responses to chat conversations. Supports streaming, tool/function calling, and various parameters to control generation behavior (temperature, max_tokens, etc.).

Why this helps

Customer feedback is hard to use when it is buried in separate calls and lengthy transcripts.

  • Groups customer problems by theme within the submitted meeting.
  • Includes supporting examples or paraphrases grounded in the transcript.
  • Helps turn qualitative calls into material for product discussions.

Setup

Build it in a few focused steps.

  • 1Connect Tldv and DeepSeek to Notis once through the portal.
  • 2Create an automation in the portal, or tell Notis in plain language what it should do.
  • 3Give one instruction asking DeepSeek to identify customer pain points, supporting evidence, and uncertainties in each Tldv transcript sent to the webhook.
  • 4Select the incoming webhook as the trigger and choose a run-report channel.
  • 5Try one real customer interview and check that each finding matches the transcript.

Questions about this workflow

Does this find trends across multiple calls?

This workflow analyzes the transcript delivered in each webhook request. Comparing trends across calls requires sending a combined set of transcripts or running a separate synthesis over collected material.

Can it quote customers?

The prompt can request short exact excerpts where available, while keeping each excerpt tied to the point it supports.

When this happens · Trigger

Do this · Action

Supported Triggers and Actions

Notis builds workflows that link Tldv to DeepSeek. A trigger fires from one place; an action lands in another.

Tldv triggers

DeepSeek actions

Recurring trigger

Notis starts this workflow on a schedule, such as daily, weekly, or during business hours.

TriggerScheduled

Create Anthropic Message

Tool to create a model response using DeepSeek's Anthropic-compatible API format with support for system prompts, tool calling, streaming, and thinking mode. Use when you need to generate chat completions using the Anthropic Messages API format. Requires base_url='https://api.deepseek.com/anthropic' or metadata base_url ending with '/anthropic'. Supports extended thinking, function calling with tools, and streaming responses. Note: Images and documents are not supported in content blocks.

ActionInstant

Webhook trigger

Notis starts this workflow when an external tool or custom backend sends an HTTP request.

TriggerInstant

Create Chat Completion

Tool to create a chat completion using DeepSeek models (deepseek-chat or deepseek-reasoner). Use when you need to generate AI responses to chat conversations. Supports streaming, tool/function calling, and various parameters to control generation behavior (temperature, max_tokens, etc.).

ActionInstant

Get User Balance

Tool to get user's current account balance including granted and topped-up balances with detailed breakdown by currency (CNY or USD). Use when you need to check if the user has sufficient balance for API calls or to view detailed balance information.

ActionInstant

List Models

Tool to list currently available DeepSeek models and provides basic information about each one such as the owner and availability. Use when you need to discover which models are available for API calls.

ActionInstant

Connect any two apps with Notis in the middle.

Tldv and DeepSeek, or any other combination from 1,000+ integrations.

When this happens · Trigger

Do this · Action

Save your first hour today.

7-day trial of any paid plan, with 20$ of usage included.
No card. Works with personal or business DeepSeek.