AI agents & agentic workflows
A multimodal quoting agent inside a messaging app
Customers send plans, photos and voice notes; the agent returns a priced quote in minutes instead of hours.
- Status
- ● Live in production
- Client
- Private
- Practice
- Engineering
Swipe to see the whole drawing →
The problem
Preparing a quote meant reading drawings, photos and material requirements by hand, often for hours. It had to work where customers already are, in a messaging app.
What we built
- A
An agent graph with a supervisor that plans the job and routes it to specialist sub-agents for each part of the quote.
- B
Messaging-app intake for text, voice, images and documents, with a queue that handles bursts of messages and only streams replies that matter.
- C
Tracing and step-level evaluation, so each change to a model or prompt is scored before release; people handle exceptions such as unavailable materials.
Stack
- Agent orchestration
- Multimodal LLMs
- Messaging API
- Tracing & evals
Designed and built by our senior engineers. The architects you'd meet on a first call are the people who built this.
Client names stay private. Every system here is real and in production, and we're glad to walk you through the ones relevant to you on a call.
Office hours
Working on something similar?
Your first call is with a senior lead who will own the work. Send a few lines about the product and what's hard about it.
- Your point of contact
- Vidit Galav, Director
- Follow
- Delivery
- Our team works across US, European, APAC and Middle East hours.