Skip to content
AI software, hardware and services//Rajshahi, Bangladesh

Rajib'sI build LLM features, AI agents and automation, RAG and search, AI mobile MVPs and local AI setups.

Practical AI work you can measure: evaluation sets, cost control and prompt-injection defenses — each with a public write-up on the blog. I also ship the web and mobile software those features sit in.

2015
Shipping software since
2017
Blockchain work since
6,375
LinkedIn followers (as of Nov 2026)
4
Live interactive demos
1

Illustrative code samples · typed live in your browser

Node.jsReactPythonLLM APIsSSE streamingpgvectorevaluation setsDjangoMapboxmobile MVPsNode.jsReactPythonLLM APIsSSE streamingpgvectorevaluation setsDjangoMapboxmobile MVPs
More demos on specialist sites: Blockchain demos · iGaming demos
Products

Ready-to-customize packages

How I usually package this work. Every engagement is scoped and quoted per project.

All products

AI & Mobile App MVP

From idea to a working web and mobile MVP with a real backend.

View package

LLM Feature Integration

Add streamed, retried, cost-aware LLM features to an existing app or a new one.

View package

RAG & Search Setup

Retrieval over your documents or records, with a small evaluation set so you can measure it.

View package
How I work

How I work: Scope → Prototype → Evaluate → Ship

  1. 01

    Scope

    We agree what the feature should do, what data it can use, and how we will tell whether it works.

  2. 02

    Prototype

    A working slice in your stack — often a React front end with a Node.js or Python API.

  3. 03

    Evaluate

    A small evaluation set and cost checks, so changes are measured, not guessed.

  4. 04

    Ship

    Handover of the working feature, with notes on retries, guardrails and what to watch in production.

> ready_to_build()

Have an AI feature in mind?

Tell me what you want to add or ship. I will reply with scope, approach and a quote.