Mar 2023 – Aug 2023 · 5 mos · Amsterdam, North Holland, Netherlands

• PROJECTS:
• 1. Automating Land Use and Land Cover Mapping Using Computer Vision and Satellite Imagery, Cameroon
• 2. Tackling Deforestation in Tanzania using Computer Vision and AI: A Pilot Project Focused on Mangroves
• ACCOMPLISHMENTS:
• 1. Orchestrated the Python-based development of deep learning models including U-Net, SegNet, DeepLab V3+, using TensorFlow, PyTorch, Keras, and Numpy, etc. with model training processes on Google Colab Pro+.
• 2. Deployed Deep Learning based Semantic Segmentation models trained on both Projects (i) Land Use & Land Cover Segmentation and (ii) Deforestation Detection in Google Cloud Platform Production Environment.
• 3. Deployed two Deep Learning models for Land Cover Segmentation and Deforestation Detection as a scalable micro-services, equipped with load balancing for handling massive traffic through horizontal scaling, on Google Cloud Platform.

Jan 2023 – Oct 2024 · 1 yr 9 mos · Amsterdam, North Holland, Netherlands · On-site

• # UNL Global has raised a Funding of $6.6 Million in 3 Rounds (including both Pre-Seed and Seed Round), from 6 Globally Recognised Venture Capital Investors including (i) Here Technologies Chicago, (ii) Elev8.VC, (iii) Venturerock, (iv) SGInnovate, (v) SOSV, and (vi) Orbit Startups, Source: CrunchBase - 2024.
• UNL - Universal Named Locations
• ACCOMPLISHMENTS:
• 1. Led the Cloud Migration of 15 Software Architectures from Google Cloud Platform (GCP) to Amazon Web Services (AWS), leveraging the three-phase migration process (Assess, Mobilize, and Migrate and Modernize) to ensure seamless transition.
• 2. Designed a Spatio-Temporal Asset Catalog (STAC), consisting of > 50 Python functions to create STAC Collection, Item, and Asset, used for cataloging and data management of different geospatial formats including .shp, .geojson, .las, .tiff, etc.
• 3. Developed an API using FastAPI in Python for the Spatio-Temporal Asset Catalog (STAC), to ensure its use across multiple services. This API service is now used for cataloging all the Geospatial data which enters the organization.
• 4. Led a Discrete Global Grid System (DGGS) grid-based analytics Project, involving creation and setting-up of multiple services as API including GeoServer (in Java), DGGS Services, ClickHouse OLAP, etc. used for global grid based analysis.
• 5. Designed and implemented a Knowledge Graph in Neo4j to model KFC’s menu for an automated drive-thru system, enabling real-time query retrieval of menu items, prices, and nutritional info using Cypher queries.
• 6. Integrated a Retrieval-Augmented Generation (RAG) model with a Knowledge Graph to process customer orders via voice input, converting them into Cypher queries for accurate menu data retrieval.

Jul 2022 – Nov 2022 · 4 mos · Enschede, Overijssel, Netherlands

• After I had Graduated and Completed my M.Sc. from the University of Twente, like every Professional I was Job Hunting and looking for a Possible Opportunity for myself in Europe, and below I would like to showcase my Failure Stories (which I feel are equally important as Success Stories) about the different companies specifically the ones in which I got invited for the interview, and the stage of my Interview which I reached, which can be seen below:
• 1. GIM (In Belgium): Rejected after Third Round
• 2. GRAS System (In Germany): Rejected after First Round
• 3. Arcadis (In Netherlands): Rejected after First Round
• 4. TAUW (In Netherlands): Rejected after Third Round
• 5. Antea Group (In Netherlands): Rejected after Second Round
• 6. SWECO (In Netherlands): Rejected after Second Round
• 7. LSC Engineering Group (In Luxembourg): Rejected after Second Round
• 8. Saxion University of Applied Sciences (In Netherlands): Selected after Second Round
• 9. UNL Global (In Netherlands): Selected after Second Round

Apr 2022 – Jun 2022 · 2 mos · The Hague, South Holland, Netherlands

• Thesis Title: Deep Learning Based Multi-Source Data Fusion to Map Deforested Areas
• ACCOMPLISHMENTS:
• 1. Designed a Multi-Modal Deep Learning Algorithm that explores Complementarity among the features from Sentinel-1 (Radar) and Sentinel-2 (Optical) data for improving Deforestation Detection using GIS.
• 2. Evaluated the Deep Learning Architecture in GIS which was prepared for Data Fusion and using standalone Sentinel-1 (Radar) and Sentinel-2 (Optical) data for both Cloudy and Cloud Free weather condition.

Jan 2022 – Mar 2022 · 2 mos · The Hague, South Holland, Netherlands

• Project Title: Above Ground Biomass (AGB) Estimation using Remote Sensing Techniques
• ACCOMPLISHMENTS:
• 1. Prepared GIS based Regression Models using different Machine Learning Algorithms in Python including (i) Decision Tree, (ii) Random Forest, (iii) Support Vector Machine (SVM), (iv) K-Nearest Neighbor and (v) ADA Boosting methods.
• 2. Estimated the Above Ground Biomass (AGB) of Trees using GIS specifically ArcMap in Tanzania in the Eastern parts of Africa using different Satellite Datasets including of Airborne LiDAR, GEDI Level-4A, Sentinel-1 (Radar) and Sentinel-2.

Apr 2022 – Jun 2022 · 2 mos · Huntsville, Alabama, United States

• Selected among the only 3 Students within the entire Batch (2020-22) of ITC, Faculty of Geo-Information Science and Earth Observations, University of Twente, to do my Internship in Collaboration between National Aeronautics and Space Administration (NASA) - SERVIR and The University of Alabama (UAH), Huntsville, United States. My Internship Topic was "Machine Learning based Fusion of LiDAR and Synthetic Aperture Radar to Generate Forest Stand Height Estimates in the Amazon". The Objective of this Project was to Prepare a Model using Python Code so as to Generate the Forest Stand Height (FSH) of entire Amazon Rain Forest using free available Synthetic Aperture Radar (SAR) and Light Detection and Ranging (LiDAR) data which can then be used in the Allometric Equation (e.g. AGB = 0.0673 × (𝜌× D2 ×H) 0.976) which can thereby be used to calculate the amount of "Carbon Sequestered" in the Amazon Rain Forest. This tool can definitely be beneficial to tackle the issue of Climate Change.
• Unfortunately, due to the restrictions as part of my University Curriculum, I was not allowed to do this Second Internship along side my M.Sc. Thesis and my First Internship which I was pursuing with a Dutch company named Space4Good. Offer Letter and Project Presentation have been uploaded in the Attachment.
• Note: SERVIR is a Department within the National Aeronautics and Space Administration (NASA), detailed information about which can be accessed at the link (https://www.nasa.gov/servir/). This specific Research Internship was a Joint Collaboration between NASA - SERVIR and The University of Alabama (UAH) in Huntsville, United States.

Sep 2021 – Jul 2022 · 10 mos · Enschede, Overijssel, Netherlands

• # University of Twente is Ranked 9th Best University in the Netherlands to Study "Machine Learning", Source: EduRank - 2024
• Machine Learning M.Sc. Thesis:
• TITLE:
• Deep Learning based Multi-Source Data Fusion to Map Deforested Areas in Amazon Rain Forest
• ABSTRACT:
• This study aims to overcome this limitation of the optical image by fusing optical (Sentinel-2) and SAR (Sentinel-1) images. We compared the accuracies of the deforestation maps generated for the year 2020 from standalone optical and SAR images with maps predicted using late fusion which includes both Sentinel-1 and Sentinel-2 sensor data as input. For cloud-free image, Sentinel-1/Sentinel-2 based late fusion provided an overall accuracy of 0.97, 0.94, and 0.91 on the full image, test set-1, and test set-2 respectively, while in the cloudy image, Sentinel-1/Sentinel-2 based late fusion provided an overall classification accuracy of 0.95, 0.91 and 0.88 respectively. In the case of a cloud-free image, the overall accuracy using Sentinel-1/Sentinel-2 based late fusion was +3%, +3%, and +3% higher for full image, test set-1, and test set-2 respectively than Sentinel-1 image and +2% and +1% higher for full image and test set-2 than Sentinel-2 image. In case of cloudy weather condition, the overall accuracy of late fusion using both Sentinel-1/Sentinel-2 image was +1%, +2% and +1% higher for full image, test set-1 and test set-2 respectively than Sentinel-1 image and +10%, +2% and +10% respectively higher than Sentinel-2 cloudy image. The presented approach using late fusion showed the advantage of fusing Sentinel-1 and Sentinel-2 sensor data for deforestation mapping compared to the standalone data source. Also, the results show significant benefits of fusing both Sentinel-1 and Sentinel-2 images even in the cloudy weather condition where 22-48% of the study area was covered with clouds in Sentinel-2 data.
• Github Repository: https://github.com/akhilchibber/Deforestation-Detection

Aug 2020 – Nov 2021 · 1 yr 3 mos · Enschede, Overijssel, Netherlands

• After I joined to pursue my Second Masters from the University of Twente, from the very first day I started hunting for Internship and a Possible Opportunity for Masters Thesis Collaboration. Below, I would like to showcase both the Success and Failure Stories (which I feel are equally important as Success Stories) as part of my Internship Hunting, about the different companies specifically the ones in which I got interviewed, and the stage of Interview which I reached, which can be seen below:
• 1. Neo BV Netherlands (In Netherlands): Selected after Second Round for Internship
• 2. Space4Good (In Netherlands): Selected after Second Round for both Internship & Thesis Collaboration
• 3. NASA (In United States): Selected after First Round for Internship
• 4. SWECO (In Netherlands): Rejected after Third Round for Internship
• 5. ESRI (In United States): Rejected after First Round for Internship
• 6. SAR Vision (In Netherlands): Selected after Second Round for Masters Thesis Collaboration

Jul 2018 – Jul 2020 · 2 yrs · New Delhi Area, India

• ACCOMPLISHMENTS:
• 1. Identified and proceeded demolition action on unauthorized construction while examining the building plan against the sanction plan on site and misuse of property as per MPD–2021 at 20 properties during Site Visits.
• 2. Automated the entire workflow of STF from manual to online by manually entry and uploading of 20,000 grievances online received in Hard-Copy with a team of 1 programmer and 3 data entry operators.
• 3. Presented am improved work-flow of STF Web-Portal along with identified issues in the existing portal to Director (Systems) resulting in approval of new tender for the STF Web-Portal with updated work-flow.
• 4. Responsible for preparing Agendas and Minutes of 35 Meetings till date held fortnightly of Special Task Force (STF) Chaired by V.C. DDA, including data analysis on action taken in daily coordination with 14 ULBs.
• 5. Completed other works including submission of 20 action taken reports to Hon’ble Supreme Court on, preparation of replies for 200 RTI matters and prepared 4 presentations till date for STF meeting.

Sep 2017 – Jul 2018 · 10 mos · New Delhi Area, India

• ACCOMPLISHMENTS:
• 1. Increased the speed of drone image processing from 5000 images per day to 20000 images per day by using multiple free trail software simultaneously including Pix4D, Drone Deploy, Sky Catch etc.
• 2. Reduced the time of QA/QC in GIS from approx. 6 hrs. to 30 minutes by automating the entire process through preparation of models using GIS model builder and also hyperlinking drone imagery as points.
• 3. Prepared Technical Proposal for the tender titled "Integrated City Development Plan" for 6 Towns in Jammu and won the project (L1) for all the 6 Towns worth consultancy fee (approx. 1.7 crore).
• 4. Completed City Development Plan, Samba as a Project Lead in a period of 2 months with responsibilities including submission of Inception, Interim and Draft Final Report and getting the payments from Client.
• 5. Managed database including survey details of approx. 37000 properties in the project titled “Improvement in Property Tax Collections for Varanasi and Lucknow” as a Team Lead of 5 GIS Experts.

Jul 2017 – Aug 2017 · 1 mo · Delhi, India

• After I had Graduated and Completed my Masters in Urban and Regional Planning from School of Planning and Architecture, Vijayawada, like every Professional I was Job Hunting and looking for a Possible Opportunities for myself in India, and below I would like to showcase my Failure Stories (which I feel are equally important as Success Stories) about the different companies specifically the ones in which I got interviewed, and the stage of my Interview which I reached, which can be seen below:
• 1. School of Planning and Architecture, New Delhi: Rejected after First Round
• 2. Malaviya National Institute of Technology (MNIT), Jaipur: Rejected after First Round
• 3. DCRUST, Haryana: Rejected after First Round
• 4. The Energy and Research Institute (TERI), Gurgaon: Rejected after First Round
• 5. NF Infratech: Rejected after Second Round
• 6. Cure India: Rejected after First Round
• 7. Spatial Decisions: Rejected after First Round
• 8. Voyants: Rejected after First Round
• 9. TERI University: Rejected after First Round
• 10. Vasudev Infraplanners: Selected after First Round
• 11. CARITAS Eco / DWI Consulting: Selected after Second Round
• I have also attached the Offer Letters, which I received from the companies in which I got selected, and also Interview Mails which I received in the companies in which I did not got selected but got a chance to give an Interview.

May 2016 – Jun 2016 · 1 mo · New Delhi Area, India

• ACCOMPLISHMENTS:
• 1. Identified gaps in facilities of Zonal Development Plan for Zone “E” under Delhi Master Plan – 2021 including green areas, electric stations, overhead tanks, police station etc. as per the URDPFI guidelines.
• 2. Drafted 14 detailed layout maps available in hard-copy in ArcMap by scanning the plan and then preparing the vector data in a short duration of one month for submission to Deputy Director.

Sep 2015 – Apr 2016 · 7 mos · Vijayawada, Andhra Pradesh, India

• After I joined my Masters in urban and Regional Planning in School of Planning and Architecture Vijayawada, from the very first day I started Hunting for Internships and the organization who selected me as an Internship Trainee Can be seen below:
• 1. Delhi Development Authority (DDA)
• 2. Rajasthan Urban Sector Internship Program
• 3. Town And Country Planning Organisation (TCPO)
• I got selected in the above mentioned three government organization for pursuing my Summer Internship, out of which I finally opted for pursuing my Internship from Delhi Development Authority (DDA). Further, I have also attached the Offer Letters, which I received from the organizations in which I got selected.

May 2014 – Jul 2014 · 2 mos · Jammu, Jammu & Kashmir, India

• ACCOMPLISHMENTS:
• 1. Learned drafting of layout plan on different software’s including AutoCAD and ArcMap and the process involved in getting the plan approved in a government organization.
• 2. Planned detailed layout and land use plan for Birpur Housing Colony, Jammu, J&K over a total area of 73 Acre with facilities as per URDPFI guidelines
• 3. Presented the final completed plan of Housing Colony to Chief Town Planner, JDA further which it was advertised by Jammu Development Authority (JDA) for implementation.