Experience
Designed and developed (work-in-progress) a software system with a web-based UI for accessing time-series data at the KAGRA gravitational wave observatory (in Gifu, Japan). This tool allows viewing of real-time data as well as previously collected data from a remote storage cluster. It could also repackage data in the selected time interval into file formats commonly used within the collaboration (.csv
, .gwf
) for downloading and offline analysis.
Below are the technologies/methods used in this project:
- Backend: Python (the AsyncIO framework, WebSocket -
aiohttp
, S3 API -boto3
, HTCondor), Docker, RESTful API, RabbitMQ, SQLite, Traefik (reverse-proxy, SSL/TLS), Proxmox, Debain. - Frontend: JavaScript, React.js, Redux.js (RTK), D3.js, MaterialUI, and CSS. Used the Vite build tool.
- Other: Used GitLab CI to automate packaging parts of the system.
- Developed a software pipeline for low-latency calibration of the KAGRA observatory data. The pipeline is utilized for the reconstruction of GW strain signal with high calibration accuracy and can be launched in both the low-latency mode for reconstruction from online data and the offline mode for reconstruction from data stored on a compute cluster. Reconstructed time-series data are used by most GW data analysis pipelines.
- Worked on calibration-related tasks: calibration model development, transfer function measurements, Photon Calibrator (Pcal) commissioning, and maintenance of the CentOS 7 server node that runs the low-latency calibration pipeline.
- Trained a new researcher to take over the Pcal system.
- Hardware activities: maintenance and improvements of the Pcal optical system and electronics, characterization of the Pcal system.
- Analysis: Wrote scripts for analyzing time-domain and frequency-domain measurements, Gaussian beam profile fitting, etc.
- Developed a Python WSGI app (HTTP) for fetching instrument data from NDS servers, doing simple transformations, and providing downsampled data and statistical summaries.
- February 2015 – December 2017, resident at the LIGO (Laser Interferometer Gravitational Wave Observatory) Hanford Observatory near Richland, Washington conducting research for Ph.D.
- Worked on calibration of Advanced LIGO detectors: measurement and analysis of the differential arm length sensing and actuation responses (took measurements, composed LTI models in Matlab, fitted model parameters).
- Developed and implemented a method for tracking and compensating for temporal variation in responses of the second-generation gravitational wave detectors.
- Analyzed calibration-related data on the LIGO Data Grid computer cluster.
- Developed methods for propagating the laser power sensor calibration from the US National Institute of Standards and Technology to the LIGO Photon Calibrator sensors.
- Research on modeling of photonic crystal nanocavities. Numerical study of the photonic band structures of models and finite-difference time-domain simulations on computer clusters at UT Rio Grande Valley and at the Texas Advanced Computing Center (TACC).
- Teaching assistant for Computational Physics, Physics, and Astronomy labs.
AlatauSoft Ltd.\ was a small IT company that offered corporate document management solutions and other systems for businesses.
- Development of a document management system on the Lotus/Domino platform for Kazakhmys (a mining corporation) and similar companies. Development of Oracle-based applications (Oracle PL/SQL, OC4J).
BSB is a company that develops and supports software and maintains some of the IT infrastructure used by the National Bank of Kazakhstan and its dependent financial institutions.
While I was studying for my B.S. in Automation, I started working at BSB on a part-time contract basis. After graduation, I switched to a full-time work.
- Development of some features and support of a subsystem called the ``Electronic Auction’’. This subsystem, developed on Delphi+Oracle PL/SQL, is one of the tools used by the National Bank for managing the currency rate by auctioning limited-time notes to private banks and agencies.