Sanzhar Myrzagalym
My projects in Data and AI
2018 — presentIn this section of my portfolio, I would like to share a few publicly available projects related to data processing and visualization, ML model training, and AI applications that I believe impacted the most to people’s lives.
In these projects, I participated as an architect, and my responsibilities included conducting proof of concepts, securing commitment from clients, orchestrating the implementation process, working with partners or Microsoft engineers.
Air Astana speeds up the reporting time by 60x and increases productivity for 100+ people
Air Astana is the flag carrier of Kazakhstan and operates both domestic and international services across 64 routes in Eurasia. The company strives for the highest safety and customer service standards for both external and internal processes. To obtain timely financial reports, the company leveraged the power of Microsoft Azure Synapse Analytics together with Microsoft Power BI. Now, reports are generated within seconds instead of hours, boosting the team’s productivity and efficiency.
Read the full story
Metinvest
achieves blast-furnace efficiency with expected added value of over $100M
With a staff of over 80,000, Ukraine-headquartered Metinvest is a major name in end-to-end steel production. Looking to improve the efficiency of its blast furnaces — reducing fuel consumption by controlling the silicon content in its iron — the company started a pilot project with Azure Data Factory and Azure Machine Learning. Having seen its efficiency improve against benchmarks set by industry leaders, the company is now looking to roll out the solution across all its furnaces.
Read the full story
Rebuilding a nation: UADamage accelerates Ukraine’s damage assessments
Half a million damaged buildings across 162 settlements in Ukraine assessed and mapped out in mere days instead of months—that's the remarkable pace at which UADamage has been able to accelerate Ukraine's reconstruction and demining efforts. Thanks to its community of engineers and a partnership with Microsoft, a city the size of Kharkiv can now be processed in just a few hours, with enhanced security for sensitive data using Microsoft Azure. UADamage not only fast-tracked damage assessments but is also working towards a future of transparent and corruption-free reconstruction.
Read the full story
Early experiments and projects
For a long time now, I have been enthusiastically studying the field of data processing and visualization, machine learning, and artificial intelligence. I have always strived to experiment and apply these skills in real-world scenarios.
Five years ago, when I had the opportunity to steer my career in this direction, I immediately seized it, and since then, I have implemented a vast number of projects using these technologies.
Here, you can find some of the examples of my explorations and experiments:
Object detection
In this experiment, I trained a basic object recognition model on an image. The percentage next to the object's name represents the model's confidence level in its recognition. The example on the right is a basic face detection model that visualizes facial landmarks, allowing us to verify the identity of the person by comparing them with another set of landmarks.
Proof-of-concept for utilizing OCR for searching through the content of project documentation (schematics)
On one of the projects, I trained a text recognition model on images to help engineers on the production floor quickly locate the necessary schematics. After implementing this project, they could quickly find documents not only by their titles and metadata but also by keywords found within the schematics themselves.