PANKAJ PRAMANIK.
LET’S TALK

02 / About

Engineering useful systems,not just impressive demos.

I’m Pankaj Kumar Pramanik, an AI and data engineer with a background in full-stack web and mobile development. I work where data, intelligent applications and everyday business processes meet.

8+
Years building software
7
Roles across global teams
20
Published projects
23
Certifications
Let’s work together Download CV
Portrait of Pankaj Kumar Pramanik, AI and data engineer
Pankaj Kumar PramanikAI · Data · Software · Automation

01 / My approach

Start with the problem.Then choose the tools.

I build document-aware AI applications, data pipelines and connected workflows - and take them past the prototype.
01

Problem first

Who needs the system, which data it can use, and what a useful result looks like - before any tool is chosen.

02

Software, not notebooks

A background in full-stack web and mobile development means AI work ends as an application people can use and maintain.

03

Reviewable steps

Architecture, prototypes and releases arrive in small pieces you can test, question and steer.

02 / Career journey

2019 → today.Building, learning, shipping.

From full-stack products and 3D web to machine learning, generative AI and RAG systems. Hover a milestone to see the stack.

Full experience
  1. 2019

    Full Stack Developer

    Anekonnect Incorporated · until 2021

    React · Next.js · Node.js · Redux-Saga · Ant Design · SQL

    • React
    • Next.js
    • Node.js
    • Redux-Saga
    • Ant Design
    • SQL
  2. 2020

    Full Stack Engineer

    Breaker Nation · until 2022

    React Native · Redux Saga · Push Notifications · App Store Deployment

    • React Native
    • Redux Saga
    • Push Notifications
    • REST APIs
  3. 2021

    3D Web Graphics & Blockchain Developer

    Vircadia · until 2022

    Solidity · Web3.js · Three.js · Babylon.js · WebGL

    • Solidity
    • Web3.js
    • Three.js
    • Babylon.js
    • WebGL
    • Blender
  4. 2022

    Full Stack Engineer

    Crewfare.com · until 2023

    Node.js · AWS Lambda · Serverless · React · Storybook

    • Node.js
    • AWS Lambda
    • Serverless Framework
    • React
    • Storybook
    • AWS EC2
  5. 2023

    Machine Learning Engineer

    Most Loved Workplace · until 2023

    Text Analytics · Statistical ML · Data Mining · Python

    • Python
    • scikit-learn
    • Text Analytics
    • Statistical ML
    • Data Mining
  6. 2024

    Generative AI & Full Stack Developer

    Subhub · until 2025

    LangChain · PineconeDB · Supabase · Twilio · Eleven Labs

    • LangChain
    • Pinecone
    • LangFlow
    • Supabase
    • PostgreSQL
    • Twilio
  7. 2025

    LLM & RAG Specialist

    Upwork · Present

    LangChain · AI Workflow Automation (n8n) · RAG · Voice & Chat Agents

    • LangChain
    • n8n
    • RAG
    • Pinecone
    • FAISS
    • OpenAI GPT-4

03 / Education

Academic foundation,applied at work.

Formal study in data science and AI alongside professional engineering work.

  1. -

    Indian Institute of Technology Guwahati

    Degree
    Bachelor of Science (Honours)
    Field of study
    Data Science and Artificial Intelligence

    Practical, industry-aligned curriculum across data engineering, MLOps, and applied AI at one of India's premier technology institutes.

  2. -

    Khulna University of Engineering and Technology

    Qualification
    Undergraduate Studies
    Field of study
    Mechanical Engineering

    Studied Mechanical Engineering before transitioning into software and technology. I later pursued a professional career in software engineering, data engineering, automation, and AI.

  3. -

    Govt. Azizul Haque College, Bogura

    Qualification
    Higher Secondary Certificate (H.S.C)

    GPA 5.00/5.00 - strongest subjects: Mathematics, Physics, and English.

04 / How I work

Clarity at every step.

Four stages, each ending in something you can see, test and decide on.
01

Understand

Define the users, constraints, available data and acceptance criteria before choosing a tool.

02

Make it tangible

Build a focused prototype so the difficult assumptions get tested early.

03

Build & evaluate

Develop in iterations, review behaviour on real inputs and document the trade-offs.

04

Launch & hand over

Prepare deployment, monitoring and operating notes so the system can be supported.

05 / Principles

What I optimise for.

Five trade-offs I make on purpose. Hover or focus a card for the reasoning.

Useful>impressive

A small system people rely on beats a demo that only works on stage.

Simple>over-engineered

Start with the least machinery that solves the problem; add complexity when it pays.

Observable>mysterious

Logs, traces and evaluations make it possible to trust - and fix - what the system does.

Production>prototype

Deployment, monitoring and handover are part of the work, not an afterthought.

Human oversight>blind automation

People approve what is costly or irreversible; automation handles the rest.

06 / Tools & capabilities

The stack behind the work.

Data Science & Machine Learning

  • Python
  • NumPy
  • Pandas
  • Matplotlib
  • Seaborn
  • scikit-learn
  • TensorFlow
  • SQL

Generative AI · LLM · Agentic AI

  • LLM (GPT-4, Claude)
  • LangChain
  • LangGraph
  • RAG Systems
  • Pinecone / FAISS
  • OpenAI API
  • n8n
  • Retell AI

Frontend

  • JavaScript
  • TypeScript
  • React
  • Next.js
  • React Native
  • Three.js
  • Redux

Backend

  • Node.js
  • FastAPI
  • Flask
  • Django
  • Serverless Framework
  • PostgreSQL
  • Supabase

Cloud & DevOps

  • AWS
  • GCP
  • Azure
  • Docker
  • Kubernetes
  • CI/CD
  • AWS Lambda
  • EC2 / S3 / RDS

Next step

Have a problem worth solving?Let’s talk.

Share what you’re working on and where it’s stuck. I’ll reply with questions, a suggested approach and whether I’m the right fit.

Let’s chat