Available for opportunities
Portrait of Mahmud Hasan Munna

Mahmud Hasan Munna

|

"Turning data into decisions — at scale."

Passionate Senior Data Scientist with 4+ years of experience building production-grade ML systems, ETL pipelines, LLM-powered applications, and analytical dashboards. I specialize in transforming raw, complex data into intelligent, scalable solutions that drive real business impact.

0+
Years Experience
0B+
Records Processed
0+
Production Projects
0
Publications
About

Who I Am

I'm Mahmud Hasan Munna, a Senior Data Scientist based in Dhaka, Bangladesh with 4+ years of experience building intelligent data systems.

My work spans the full data science lifecycle — from raw ETL pipelines processing billions of records, to deploying LLM-powered applications that redefine how organizations interact with their data. I've built production systems at fintech and payment gateway companies, delivering measurable business impact at every step.

With a research background (3 international publications), a RUET engineering degree, and hands-on experience with the modern ML stack, I bridge the gap between academic rigor and real-world execution.

LLM & Agentic AI

Building intelligent agents and RAG pipelines using LangChain, LangGraph, and cutting-edge LLMs.

Data Engineering

Designing scalable ETL pipelines, data warehouses, and real-time dashboards for billion-record datasets.

ML & MLOps

Deploying production-grade ML models with Docker, FastAPI, CI/CD, and cloud infrastructure on AWS.

GitHub Activity

Contribution history · last 12 months

 
Career

Professional Experience

Building impactful data solutions across fintech and payment industries.

Wegro Technologies Limited

Current

Senior Data Scientist

Leading data science initiatives for an Agri-Fintech platform, building end-to-end ML pipelines and intelligent automation systems.

  • Built a company-wide data warehouse and automated ETL pipelines.
  • Developed eKYC onboarding module, making farmer onboarding 90% faster and eliminating fake registrations.
  • Created a BI chatbot using LLMs, reducing surface-level analytics time by ~70%.
  • Automated 30+ recurring financial and operational reports, saving ~80 hours of manual work per month.
  • Delivered production-ready scoring models and dashboards, improving decision-making efficiency by ~50%.
PythonLLMsETLPostgreSQLAWSDockerFastAPI

SSL Wireless

Data Scientist

Worked on large-scale data pipelines and ML solutions for one of Bangladesh's leading payment gateway companies.

  • Processed 1B+ customer behavior records using Dask and Airflow to generate targeted marketing insights for large Payment Gateway Clients.
  • Built an NLP-based SMS Compliance Validator with 95% accuracy, deployed to process 100,000+ messages per hour.
  • Developed geospatial fraud detection solution used by 300+ field agents, improving field audit efficiency.
  • Built and deployed forecasting and analytics pipelines that reduced recharge balance-out incidents for telecom operations.
DaskAirflowNLPGeospatialFastAPIDockerCI/CD
Work

Featured Projects

Production systems built at scale — from LLM pipelines to billion-record ETL workflows.

Featured
LLM / Agentic AI

Business Intelligence ChatBot

Agentic AI system that queries databases using natural language prompts, reducing analytics time by ~70% across the organization.

Agentic AILLMsLangChainLangGraphPostgreSQL+4
Featured
Computer Vision

eKYC Farmer Onboarding Module

Computer Vision + LLM pipeline for accurate identity verification, making farmer onboarding 90% faster and eliminating fake registrations.

Computer VisionOCRLLMFastAPIDocker+4
Featured
Data Engineering

Centralized Data Warehouse (Agri-Fintech)

End-to-end data warehouse solution with automated ETL pipelines, analytics dashboards, and real-time reporting for an agricultural fintech company.

ETLPostgreSQLPandasAirflowAWS EC2+4
Data Engineering

Billion-Scale Customer Behavior ETL

ETL pipeline processing 1B+ customer behavior records for targeted marketing and credit scoring for large payment gateway clients.

ETLMySQLDaskAirflowPlotly-Dash+4
LLM / Agentic AI

Terms & Conditions Validator (RAG + LLM)

RAG-based system for validating merchant terms and conditions for SSLCommerze, improving compliance checking accuracy and speed.

LLMsLangChainRAGFAISSAWS EC2+4
Machine Learning

Geospatial Fraud Detection (FMCG)

ML-powered fraud detection for sales representatives using geospatial analysis of supply chain app data, used by 300+ field agents.

Geospatial AnalysisTraditional MLFastAPIDockerCI/CD
Machine Learning

Telecom Recharge Transaction Forecasting

Time series forecasting pipeline for telecom operator recharges through Easy.com, reducing balance-out incidents significantly.

Time Series AnalysisNeural ProphetFastAPIDockerCI/CD
LLM / Agentic AI

Automated Meeting Minutes Summarizer

Speech-to-text + LLM pipeline that transcribes meetings, separates dialog speakers, and generates structured meeting minutes automatically.

LLMsNLPWhisperDialoGPTDocker
Data Engineering

Automated Financial Report ETL & Bot

Automated accounting and financial report generation with ETL pipelines, visualization dashboards, and a report distribution bot.

ETLPostgreSQLPandasNumpyAirflow+5
LLM / Agentic AI

Data Pipeline from Messy Doccument

Developed an end-to-end AI-powered data pipeline to process and analyze over 500 unstructured PDF documents, each exceeding 20 pages in length. Leveraging OCR and Large Language Models (LLMs), the solution automatically extracted more than 25 key variables from complex document layouts and converted them into structured, analysis-ready datasets.

PythonLangChainOCR (AWS Textract/Tesseract)PandasPostgreSQL+2
Expertise

Skills & Technologies

A comprehensive toolkit spanning the full ML and data engineering stack.

Programming Languages

PythonSQLCMATLAB

ETL & Analytics

PandasDaskPolarsNumPyApache AirflowPlotlySeabornStreamlitPlotly-DashLooker Studio

Machine Learning

Scikit-learnDNNCNNRNNTensorFlowPyTorchTransformersNeural Prophet

LLMs & Generative AI

LangChainLangGraphVector DatabasesFAISSRAGPrompt EngineeringAgentic AI

Computer Vision

OpenCVOCRTransformersImage Processing

Deployment & MLOps

DockerFastAPIGitHub Actions (CI/CD)OAuth2JWTAWS EC2AWS RDSAWS S3
Research

Publications

Peer-reviewed research contributions in machine learning and NLP.

Sentiment Analysis and Product Review Classification in E-commerce Platform

Authors: M. H. Munna, M. R. I. Rifat and A. S. M. Badrudduza

2020 23rd International Conference on Computer and Information Technology (ICCIT)Dhaka, Bangladesh

NLPSentiment AnalysisE-Commerce
DOI: 10.1109/ICCIT51783.2020.9392710

An End-to-end Machine Learning System for Mitigating Checkout Abandonment in E-Commerce

Authors: M. R. Islam Rifat, M. Nur Amin, M. H. Munna and A. Al Imran

2022 17th Conference on Computer Science and Intelligence Systems (FedCSIS)Sofia, Bulgaria

Machine LearningE-CommercePrediction
DOI: 10.15439/2022F167

Identification of Clickbait in Video Sharing Platforms

Authors: M. H. Munna and M. S. Hossen

2021 International Conference on Automation, Control and Mechatronics for Industry 4.0 (ACMI)Rajshahi, Bangladesh

NLPClickbait DetectionDeep Learning
DOI: 10.1109/ACMI53878.2021.9528095
Background

Education & Certifications

Academic foundation in engineering, complemented by industry-recognized training.

Academic Education

BSc in Electronics and Telecommunication Engineering

Rajshahi University of Engineering and Technology (RUET)

Rajshahi, Bangladesh·

Higher Secondary Certificate (HSC)

New Govt. Degree College

Rajshahi, Bangladesh·

Secondary School Certificate (SSC)

Mohanpur Govt. High School

Rajshahi, Bangladesh·

Certifications & Training

Training on Artificial Intelligence

PUM Netherlands Senior Experts, Organized by BASIS

Cloud Journey With AWS

BASIS

Milestones

Achievements & Activities

Beyond code — recognition, competition, and community.

Achievements

Robi Datathon 3.0 Grand Finalist

Achieved Grand Finalist position in one of Bangladesh's most prestigious data science competitions, organized by Robi Axiata Ltd.

2024

Extra Curricular

  • Opening Batsman at Inter Software Company Cricket Tournament (ISCCT) for SSL Wireless
  • Former Member and Batsman, Cricket Club of RUET (CCR)
  • Sony Smart Karaoke Music Competition - Top 200 Nationwide
Contact

Let's Work Together

Open to new opportunities, collaborations, and interesting data problems.

Got a project in mind?

Whether you're looking for a Senior Data Scientist, need help with ML systems, LLM pipelines, or data infrastructure — I'd love to hear from you.

Send me an email

Location

Banasree, Rampura, Dhaka, Bangladesh