We build AI products that make it to production.

We're a small team in Dhaka focused on AI agents, LLM apps and the software around them, and we take projects from first prototype to live product. You talk directly to the engineers who write your code, not to an account manager or a ticket queue.

We've shipped in marketing tech (Creative Genie, Creative Xperience) and contract analysis (Comvado); we take on other industries where the problem fits. We work with startups that need a first version and with established companies adding AI to systems they already run, and we scope each project to your stage and budget.

We build LLM integrations (OpenAI, Claude, Gemini), AI agents that carry out multi-step tasks in your tools, assistants that answer from your own documents, document processing and recommendation features. We set up deployment and monitoring as part of the work, so you can see how it behaves in production.

Both. We add AI features to existing software without rebuilding it, and we work on existing codebases that have become slow or hard to change. We roll changes out in steps so your users aren't disrupted.

Mostly Python (Django, FastAPI) on the backend, Next.js and React on the frontend, PostgreSQL and Redis for data, and Docker-based deployments on AWS or similar clouds. We pick what fits your project and document the decisions.

It depends on scope. After the first fixed-scope step we give you a milestone plan with dates before the build starts, and you see working software at the end of every sprint.

Yes. Once the work is paid for, the source code, design files and documentation are yours, together with access to the repositories, credentials and admin accounts. We're happy to sign an NDA before you share any documents, data or system access.

Ongoing maintenance scoped to what you need: bug fixes, security updates, monitoring, performance tuning and new features.

We plan for it from the first release: encrypted connections, secure authentication, least-privilege access and checks against common vulnerabilities such as the OWASP Top 10. For personal data we follow GDPR principles: collect only what is needed, make it possible to delete it and log who accessed it.

Yes. We sign a data processing agreement with the EU Standard Contractual Clauses on request, and we work inside your own cloud account and repositories.

Yes. We can work under your standard contract and your country's law, after review.

We invoice in EUR or USD.

Our team works from Dhaka (UTC+6). We share the European morning every working day. For US clients we agree one fixed daily call time at kickoff.