01
New projects
A paragraph about the problem is enough to start. We reply within two working days, usually with questions rather than a quote.
AI · Service Generative AI and chatbot development
Chatbots and generative AI features grounded in your own data, connected to your systems, measured with evaluations and kept inside a cost budget.
In short
Generative AI development means building features that write, summarise, classify or answer using large language models. A chatbot is the most familiar form: a conversation that answers questions and completes tasks. We build chatbots and generative features that answer from your own content, hand over to a person when they should, and are tested against real conversations before customers use them.
- How to engage
- Product build, Dedicated engineering pod
- Practice
- AI
Signs this is the right service
- Customers or staff ask the same questions every day and the answers already exist in your documents.
- Support, sales or operations teams spend hours summarising, drafting or classifying text.
- A chatbot exists but gives generic or wrong answers and nobody trusts it.
- You want AI inside your product, not a separate tool your users have to open.
What you get
Use-case selection
The tasks where generative AI saves real time, ranked by value and risk, so the first release is one that can succeed.
Grounded answers
Responses drawn from your knowledge base, tickets and documents through retrieval, with sources shown to the user.
Channels
Web chat, your product, WhatsApp, Slack or Teams — wherever your users already are — with the same answers in each.
Handover to people
Clear rules for when the bot passes a conversation to a person, with the context attached so nobody asks twice.
Evaluation and cost control
Tested against real past conversations, with traces, feedback capture and a spending ceiling in production.
The process
- 01
Pick the use case
One job with a clear success measure, chosen from the conversations and tasks you already have.
- 02
Collect the knowledge
Connect the documents and systems the answers come from, with permissions intact.
- 03
Build and evaluate
A working assistant scored against real examples before anyone outside the team sees it.
- 04
Launch and learn
Release to a share of users, review the conversations it struggles with, and improve.
ClaudeOpenAIRAGpgvectorTypeScriptPythonWhatsApp Business API
Generative AI & chatbots: common questions
What is the difference between a chatbot and an AI agent?
- A chatbot mainly answers questions in a conversation. An agent takes actions — it calls tools, updates records and completes tasks. Many projects start as a chatbot and add agent capabilities once the answers are trusted.
Can the chatbot answer from our own documents?
- Yes. That is the main point of building one: it retrieves the relevant passages from your content and answers from them, showing its sources.
Can it work on WhatsApp?
- Yes, through the WhatsApp Business API, as well as on your website, in your app or in internal tools such as Slack and Teams.
How do you stop it making things up?
- By answering only from retrieved content, saying "I do not know" when nothing relevant is found, and testing against real questions before launch.
06 / Let’s build
Bring us the difficult problem.
02
Careers
Engineers, designers and researchers who like systems that are hard to get right. Send something you built and what you would change about it now.
03
Studio
Based in India and working remotely, with overlap hours agreed up front so your team always has a time to reach us.
04
Read
Guides on AI agents, RAG, MCP, tokenization and chain abstraction, written by the engineers who build them.