Case study
AI agents for company research.
Brandy is an internal Clait project: starting from a company’s website, a multi-agent system gathers information from public sources, structures it into a report and drafts strategies and first-contact messages, supporting lead generation and prospecting.
Published
01
The project
Brandy started inside Clait as an application of agentic systems to data research. When an agency signs up, it reads the agency’s website and documents to learn its profile and services.
For each company to approach, the user enters its name, website, any instructions and the language: Brandy starts the research and shows its progress live.
02
How the agents work
Brand identity
Analyses the company’s website, up to 15 pages.
First research
Gathers the basics from the web and Google Places.
Supervisor
A supervisor agent decides which line of research to deepen next, for up to 12 rounds: OSINT, general research, Meta ads, e-commerce, online reputation, data checks.
Consolidation
Everything is merged into a narrative report and an objective report.
Consistency check
A final check reduces the risk of claims the sources do not support.
03
The sources
Web search engines, Google Places, the Meta Ad Library, review platforms such as Trustpilot, Glassdoor and Indeed, and presence on the main marketplaces. On the agency’s side, Brandy can also draw on the HubSpot CRM to know the clients already won.
04
What it produces
Company report
Profile, market position, products and services, decision makers, competitors, digital presence and online reputation.
Indicators
Digital maturity from 1 to 10, likelihood of engagement from 0 to 100 and online reputation from 1 to 5.
Strategies
Six tailored strategies, each with an estimated budget and hours of work.
First contact
Email drafts, a phone script and LinkedIn or Instagram messages.
Report chat
A chat to refine the report, including with fresh research.
05
Technology
- Agents
- LangGraph, LangChain
- LLM
- OpenAI GPT-5.4 and GPT-5.4 mini, open models on Groq
- Backend
- Python, Django, Django REST Framework, Celery, Redis
- Data
- PostgreSQL
- Frontend
- Vue 3, TypeScript, Tailwind CSS
- Observability
- Langfuse
FAQ
Frequently asked questions.
What is Brandy?
Brandy is an internal Clait project: a multi-agent system that researches a company across public sources and prepares reports, strategies and first-contact drafts.
Which sources does it use?
The company’s website, search engines, Google Places, the Meta Ad Library, review platforms such as Trustpilot, Glassdoor and Indeed, and marketplaces.
Does Brandy write the first contact too?
Yes: it drafts emails, phone scripts and LinkedIn or Instagram messages, which the user reviews before using them.
Keep reading.
- Service · 01AI agentic systems for complex processesClait develops LLM-based agentic systems: specialised AI agents that work together to orchestrate complex workflows, make decisions, correct themselves and work with the data and tools a company already uses.Agentic systems · Works with your tools · Automation
- Case study · 02Querying a database in natural languageClait built a conversational business intelligence platform: staff type a question in plain language, the system turns it into an optimised SQL query and returns an interactive chart, with no SQL knowledge needed.Agentic systems · LLM · SQL
Contact
We find a solution for everyone.
We'll find yours too.
Tell us what you'd like to improve, even if for now it's only an idea. We'll start from there and build the rest together.