Tokyo, Japan / Geospatial Big Data Engineer

Bikash Sapkota

Building data platforms for smart cities and urban planning.

I build production data systems across geospatial analytics, AI/ML, and optimization. My work spans mobility ETL/ELT pipelines, people-flow analysis, pricing optimization, OpenADR integrations, and OCR-driven document intelligence.

Contact by emailBook a meetingView GitHub

Experience

Recent work across geospatial data, AI systems, and applied optimization.

The through-line is production work: turning raw data, models, and business logic into systems people can use.

Jan 2025 - Present

Big Data Engineer (GeoSpatial)

LocationMind / Permanent

Tokyo, Japan

Amazon EC2Data AnalysisETL/ELTGeospatial joinsTime-series features
  • Design, build, and maintain scalable ingestion and ETL/ELT pipelines for mobility and geospatial datasets with reproducible data quality.
  • Implement validation, normalization, aggregation, geospatial joins, and time-series feature generation to make location data analysis-ready.
  • Develop analytical models and outputs for people-flow analysis, visualization, dashboards, maps, and CSV data products.
  • Partner with researchers, consultants, and product teams to support mobility use cases using people-flow focused AI models and features.
  • Improve monitoring, performance, cost, documentation, and governance across the data lifecycle.

Nov 2021 - Jan 2025

AI Engineer

GridSolutions Inc / Permanent

Tokyo, Japan

Machine LearningPythonOptimizationMINLPOpenADRAPI design
  • Researched electricity pricing problems and implemented core business logic for optimization workflows.
  • Created and deployed optimization modules using Mixed-Integer Nonlinear Programming to improve pricing strategies.
  • Developed API interfaces for integration with client platforms.
  • Deployed, maintained, and resolved bugs in OpenADR VEN projects.
  • Refactored project architecture to remove single points of failure.

Nov 2019 - Nov 2021

Machine Learning Engineer

Bottle / Permanent

Kathmandu, Nepal

AWS LambdaAmazon EC2Machine LearningREST APIsTeam leadership
  • Led end-to-end project lifecycles as a team lead, from planning through delivery.
  • Collaborated with system architects to define implementation architecture aligned with client needs.
  • Developed RESTful APIs that integrated machine learning functionality into client applications.

Jul 2018 - Nov 2019

Machine Learning and OCR Engineer

Smart Data Solutions / Full-time

Kathmandu, Nepal

OCRTesseractFineReaderWEKARandom ForestNER
  • Extracted characters from scanned paper claims using Tesseract, FineReader, and Cartouche.
  • Classified claims from extracted text using Random Forest models in WEKA.
  • Extracted information from claims using regular expressions and named entity recognition.
  • Improved manual keying interfaces to speed up operational workflows.

Education

Academic foundation in computer science and information technology.

Formal computer science training that supports the later career arc across software, data systems, machine learning, and cloud applications.

Nepal

BSc CSIT

Tribhuvan University

Bachelor of Science in Computer Science and Information Technology, grounding later work in software systems, data, and applied machine learning.

Skills

A practical stack for data-heavy systems.

Grouped from production roles across geospatial analytics, machine learning, cloud services, and delivery.

Geospatial Data

Mobility datasetsPeople-flow analyticsGeospatial joinsLocation intelligenceMap-ready outputs

Data Engineering

ETL/ELT pipelinesValidationNormalizationAggregationData qualityGovernance

AI, ML, Optimization

Machine learningMINLPPricing optimizationRandom ForestNamed entity recognitionOCR

Backend and Cloud

PythonREST APIsAWS LambdaAmazon EC2OpenADR VENCloud applications

Delivery

Stakeholder deliverablesDashboardsCSV data productsMonitoringPerformance tuningCost optimization

Services

Technical services around data, AI, and cloud applications.

Useful for collaboration, technical discussion, and projects where data systems need to become reliable products.

Geospatial data platforms

Scalable ingestion, preprocessing, enrichment, and analysis workflows for mobility and location datasets.

AI/ML systems

Production-oriented machine learning integrations, feature pipelines, and applied model workflows.

Optimization workflows

Business logic and optimization modules for pricing, decision support, and system integration.

Cloud data applications

APIs, dashboards, exports, and cloud-hosted services that make analytical systems usable by teams.

Case studies

Selected examples from recent company work.

A concise view of technical scope, outcomes, and delivery patterns across geospatial data, analytics, and optimization systems.

LocationMind / Geospatial data engineering

Mobility Data Platform

Built ingestion and transformation workflows that turn raw mobility datasets into reusable analytical foundations.

  • Standardized raw location inputs into reusable analytical datasets.
  • Added quality checks, normalization, aggregation, and geospatial enrichment.
  • Produced data products ready for analysis, visualization, and downstream modeling.
ETL/ELTGeospatialData qualityTime series

LocationMind / Location intelligence products

People-Flow Analytics Outputs

Turned movement data into stakeholder-ready analytical outputs for research, consulting, and product workflows.

  • Modeled people-flow patterns for dashboards, maps, and CSV exports.
  • Supported research, consulting, and product workflows with reproducible data outputs.
  • Connected data engineering work to AI-assisted mobility use cases.
People flowDashboardsMapsData products

GridSolutions / AI and optimization systems

Energy Pricing Optimization

Implemented optimization and integration workflows for electricity pricing systems.

  • Implemented optimization modules using Mixed-Integer Nonlinear Programming.
  • Built API interfaces for client-platform integration.
  • Improved reliability by refactoring architecture and maintaining OpenADR VEN projects.
MINLPPythonAPIsOpenADR

Contact

Open to collaboration and technical discussion.

Reach out about geospatial data platforms, AI/ML systems, optimization workflows, or cloud data applications.

[email protected]Book a meeting

Profiles

GitHubLinkedInFacebook

Details

Tokyo, JapanBSc CSIT, Tribhuvan UniversityEnglish / Nepali