01 / CORE TRACK
Data Analytics & Engineering
Data collection, cleaning, transformation, validation, statistical analysis, and repeatable pipelines.
Selected evidenceGeo-business data pipeline · Climate-data validation
Explore projectsCOSMIC RESEARCH NODE · DHAKA / EARTH
NELOY PRAMANIKSUPTODATA ANALYTICS · MACHINE LEARNING · NETWORK ENGINEERING
Data Analytics · Machine Learning · Network Engineering
I work across data analytics, machine learning, and network engineering—turning datasets, research questions, and infrastructure requirements into testable technical solutions.
ATOM / ML
SIGNAL / DATA
ORBIT / NETWORKS
DEEP SIGNALACTIVERESEARCH LINK
MACHINE LEARNINGDATA ANALYTICSEXPLAINABLE AIDATA ENGINEERINGNETWORK ENGINEERINGRESEARCH SYSTEMS
THREE CORE DISCIPLINES / EQUAL PRIORITY
Data. Intelligence. Infrastructure.
Three interconnected career paths, supported by hands-on projects, technical training, and ongoing research.
02 / CORE TRACK
Machine Learning & AI
Supervised learning, NLP, explainability, careful benchmarking, and research-driven prototypes.
Selected evidenceSleep health ML · Bengali XAI · Drought research
Explore projects03 / CORE TRACK
Network Engineering
IP network design, routing, segmentation, security fundamentals, and virtual network labs.
Selected evidenceMulti-branch network lab · Cisco / MikroTik training
Explore projects01Identity scan
Research-minded work across data, AI, and networking.
Three complementary technical interests, grounded in reproducible experiments, data workflows, and practical network engineering.
HUMAN / NPSMirpur, Dhaka, Bangladesh
I am a Computer Science and Engineering graduate from Daffodil International University. My three core professional interests are data analytics and engineering, machine learning and AI, and computer networking.
My hands-on work includes data acquisition and validation pipelines, explainable ML research, and simulated enterprise networks using Cisco and MikroTik concepts. I value reliable evaluation, practical engineering, and clear technical documentation.
EducationB.Sc. in CSEDaffodil International University · CGPA 3.60 / 4.00
Current orbitData Engineering AssociateData Operations & Analytics · Part-time / Remote
Research statusPublished · Accepted · Under reviewPublication status and manuscript stage are shown separately below.
02Capability solar system
Three core skill tracks. One connected toolkit.
Data analytics, machine learning, and networking come first. Deployment and programming support all three disciplines.
ORBITAL CAPABILITY MAPLIVE SYSTEM
An interactive solar system illustrates the connected toolkit supporting data, machine learning, and networking.
01
Data Analytics & Engineering
PythonSQLPandasNumPySciPystatsmodelsData CleaningFeature EngineeringEDAStatistical TestingPower BIExcelPower QueryMatplotlib
02
Machine Learning & AI
Scikit-learnTensorFlowKerasXGBoostLightGBMCNNLSTMNLPBanglaBERTmBERTSHAPLIME
03
Networking & Systems
Cisco IOSMikroTik RouterOS v7OSPFRIPVLANInter-VLAN RoutingDHCPNATACLVRRPSSHv2SyslogNTPEVE-NGCisco Packet TracerVirtualBox / VMware
04
ML Engineering & Deployment
FastAPIMongoDBDockerAWS EC2AWS ECRGitHub ActionsGit/GitHubLinuxJupyter Notebook
05
Programming & Automation
PythonSQLC (basic)JavaScriptNode.jsPlaywrightCSV processing
03Mission archive
Explore work across data, ML, and networks.
Filter the six featured projects by technical area. Each project links to a detailed case study and its source repository.
Showing 6 of 6 featured projects
MISSION01
Sleep Health Prediction System — End-to-End ML Pipeline
An end-to-end machine learning system for predicting sleep-related health conditions from lifestyle, behavioral, and medical features using a self-collected dataset.
Pythonscikit-learnFastAPIMongoDBDockerAWS EC2
MISSION02
Agro-Climatic Data Reconstruction with Machine Learning — BD-AgroClim
A reproducible research workflow comparing NASA POWER and ERA5-Land against 36 BMD stations and correcting daily agro-climatic reanalysis bias with machine learning.
PythonPandasNumPyscikit-learnXGBoostLightGBM
MISSION03
Bengali Fake-News & Hate-Speech Detection with Explainable AI
A controlled comparison of BanglaBERT and multilingual BERT on Bengali fake-news and hate-speech tasks, paired with statistical testing and quantitative explanation evaluation.
PythonBanglaBERTmBERTTransformersscikit-learnSHAP
MISSION04
Enterprise Multi-Branch Network with High Availability & Multi-Area Routing
A simulated multi-site enterprise network built in EVE-NG with MikroTik RouterOS v7, multi-area OSPF, segmented VLANs, security controls, and gateway high availability.
MikroTik RouterOS v7EVE-NGOSPFVLANVRRPNAT
MISSION05
Enterprise Geo-Business Discovery & Data Collection Pipeline
A modular Node.js and Playwright workflow that transforms publicly available Google Maps business results into structured CSV datasets.
Node.jsPlaywrightJavaScriptCSVParallel ProcessingData Validation
MISSION06
Agenting AI — Reflection & Memory Prototype
An experimental notebook-based prototype exploring reflection-driven learning, structured error memory, and iterative reasoning for self-improving AI workflows.
PythonJupyter NotebookJSONPrompt EngineeringReflection LogicMemory Systems
More engineering work is available in the full GitHub profile, including a MERN-based cooking & kitchen product store application.
View all repositories04Deep signal records
Research, with its status clearly stated.
Peer-reviewed publication, accepted-but-unpublished work, accessible preprints, and ongoing manuscripts in ML, environmental data, and climate forecasting.
DEEP SPACE RECEIVERLIVE
Published, accepted & available preprints
Publication status is stated separately from public preprint availability.
01
Addressing Behavioral Patterns of Late Sleepers Using a Supervised Learning Approach
IDAA 2025 conference proceedings · Atlantis Press
A supervised-learning study of late-sleep behavior and associated behavioral patterns.
View publication02
An Explainable Machine Learning Framework for Personalized Sleep Health Assessment and Behavioral Recommendations
Indonesian Journal of Electrical Engineering and Computer Science (IJEECS)
An explainable machine-learning framework for sleep-health assessment and behavior-oriented recommendations.
Accepted on 10 September 2026 (paper ID 46266); publication was not completed. The linked SSRN version is a preprint, not the journal publication.
Read available preprint03
Hybrid CNN–LSTM Framework for Long-Term Solar Irradiance Prediction Using Multi-Source Satellite Data: A Spatio-Temporal Analysis over Bangladesh
SSRN preprint
Multi-source satellite climate data and a hybrid deep-learning framework for solar irradiance analysis and prediction across Bangladesh.
Read preprintManuscripts under review
Research summaries only · Journal names and full manuscripts are not publicly displayed.
04
When does daily reanalysis bias correction transfer? Spatial separation, environmental support and climate-index trade-offs
Journal manuscript · Under review
Evaluates ERA5-Land bias correction across 36 Bangladesh weather stations, testing spatial transfer, training budgets, and climate-index trade-offs. Under station withholding, local Extra Trees reduced maximum-temperature RMSE by 40.9%; the gain was 34.3% with a 100-km exclusion buffer.
Manuscript results are under peer review; journal identity is intentionally not listed.
05
Observation-calendar and meteorological-source sensitivity in drought-direction diagnostics across Bangladesh
Journal manuscript · Under review
Investigates how missing-observation calendars, preprocessing, and meteorological data sources influence directional drought diagnostics at 36 Bangladesh locations, using CCM, Granger tests, PCMCI, and controlled simulations.
Manuscript results are under peer review; journal identity is intentionally not listed.
06
Apparent Drought-Forecast Skill Tracks Target-Window Overlap: A Multi-Scale, Target-Date-Embargoed Benchmark across Bangladesh
Journal manuscript · Under review
A 24-task drought-forecast benchmark across 36 stations. Overlap-aware baselines explained much of the apparent skill; tested ML models did not beat the validation-selected baseline on any of the 24 tasks.
Neloy Pramanik Supto is an equal-contributing second author and corresponding author. Manuscript results are under peer review; journal identity is not listed.
Under review does not mean accepted or published. Results described above are from submitted manuscripts and remain subject to peer review.
05Flight log
From CSE foundations to applied data operations and research.
Experience orbit
Mar 2026 — Present
Data Engineering Associate (Data Operations & Analytics)
Red Digital Marketing Group
- Conduct data acquisition, preprocessing, cleaning, transformation, and validation workflows supporting marketing analytics and business insights.
- Convert raw and unstructured data into structured, analysis-ready datasets using consistency checks and quality-control rules.
- Organize datasets with basic data modeling to improve reporting and downstream analytics usability.
Education orbit
2021 — 2025
B.Sc. in Computer Science and Engineering (CSE)
Daffodil International University
- Relevant study areas include Data Mining & Machine Learning, Artificial Intelligence, Natural Language Processing, Statistics & Probability, Database Management Systems, Data Structures & Algorithms, Computer Networks, Data Communication, Information Security, Operating Systems, and Big Data & IoT.
06 / PROFESSIONAL DOCUMENTS
One background. Three tailored CVs.
Choose the document that best matches the role. All three PDF downloads are the original versions supplied by the candidate.
TWO-PAGE PDFDATA / ML
Data & Machine Learning
Data analyst, data engineering, applied ML and research-oriented roles.
TWO-PAGE PDFNETWORK ENGINEERING
Networking & NOC
Networking, NOC, network support, Cisco / MikroTik and infrastructure roles.
TWO-PAGE PDFGENERAL / CROSS-FUNCTIONAL
Overall Technical Profile
A combined CSE profile for broader junior technical and engineering roles.
07 / CREDENTIALS / PROOF OF LEARNING
Courses, training & research participation.
View the actual certificate images and supplied evidence links. Training completion and conference participation are labelled accurately rather than presented as vendor exams.
Showing 8 of 8 supplied certificate records. Evidence links open externally.
DATA · 2026
Data Analytics with Excel, Power BI, Python & SQL
Tutorials Point Bangladesh
Course completion · Certificate ID 26925b25c5ae5683
View supplied certificate
DATA · 2026
Introduction to Data Engineering
IBM · Coursera
Course completion · Coursera ID WDYVTAPD8EYC
View supplied certificate
ML · 2026
Machine Learning Using Python
Simplilearn SkillUp
Course completion · Certificate code 9686515
View supplied certificate
NETWORKING · 2026
Networking Basics
Cisco Networking Academy
Cisco Networking Academy course completion
View supplied certificate
NETWORKING · 2026
CCNA 200-301 & MTCNA Training
Creative IT Institute
Grade A course completion · Issued April 2026; training completed July 2025. Not a Cisco/MikroTik vendor exam credential.
View supplied certificate
RESEARCH · 2025
IDAA 2025 — Paper Presentation
International Conference on Intelligent Data Analysis and Applications
Certificate of appreciation · Paper ID 10381 · Accepted and presented
View supplied certificate
OTHER · 2026
AI Product Management Micro-Certification
Product Space
Certificate of participation (not an examination credential)
View supplied certificate
OTHER · 2025
Passive Income & Content Creation with AI
Learning Bangladesh
Course completion · Certificate ID 97
View supplied certificateAdditional training
Other coursework listed in the CVs; individual evidence links have not yet been supplied.
Data Science Essentials with Python · Data Analytics EssentialsCisco Networking Academy
Python Essentials 2Python Institute
AI FundamentalsIBM SkillsBuild
Linux EssentialsNDG / LPI
Certificates are displayed as course completion, conference participation or training evidence as stated on the documents. A shared Google Drive link is not, by itself, an issuer-verification service.
08Collaboration modules
Three practical ways I can contribute.
Grounded in the projects, coursework, and current data-operations experience shown above.
01
Data Analysis & Automation
Data acquisition, cleaning, validation, exploratory analysis, structured reporting, and repeatable automation workflows using Python, SQL, spreadsheets, Power BI, Node.js, and Playwright.
02
Machine Learning Prototyping
Applied ML workflows spanning preprocessing, feature engineering, model comparison, explainability, evaluation, and lightweight API deployment.
03
Network Lab & Topology Design
Simulated Cisco and MikroTik network labs covering IP planning, VLANs, OSPF, DHCP, NAT, ACLs, VRRP, troubleshooting, logging, and high-availability concepts.
09 / OPEN COMMUNICATION CHANNEL
Have a role, research idea, or technical project in mind?
Send a message from the site, or reach out through email, LinkedIn, or GitHub.
neloybbs@gmail.com Mirpur, Dhaka, Bangladesh