He is a TESS postdoctoral associate at MIT with a visiting appointment at Princeton University. He works on the discovery and characterization of planets beyond our Solar System and detection of dark matter, where he builds novel inference tools using Bayesian statistics and machine learning. He is a group vetting lead of NASA's Transiting Exoplanet Survey Satellite (TESS) and works on many aspects of the analysis of TESS data.
- First of all, can you tell us a little about yourself?
Of course, I work as a postdoctoral researcher. My research area is exoplanets. We generally do numerical projects, using datasets to test scientific hypotheses and make inferences and work on various problems on exoplanets and dark matter. Discovery and characterization is also my area of indirect exploration on dark matter.
- Secondly, how did you draw your career path? So how do you evaluate the steps you have taken so far?
I actually had a plan to pursue an academic career from the very beginning. The reason why I turned to this field is probably my love and interest in the night sky. I had a particle physics adventure, I studied particle physics for 3 years, I went to summer school at CERN during my undergraduate period. Then I did a few projects, then I worked on astrophysical problems at the doctoral level. In other words, I can say that I was planning this road from the very beginning, and I did not encounter many surprises.
- What research skills do you think you have gained during your academic or research career?
More numerical methods, data processing, data visualization and coding. If 90 percent of my work is coding and 10 percent is simple mathematical modeling on paper, sometimes integration, but I do not use very heavy mathematical methodology and symbology. Mostly numerical. Since a small part of my work is astronomical, using a telescope or something, but I am not an observer, so I did not specialize in that part.
- Can you tell us about the research methods you used in your project?
As I said, we are actually trying to create a data base in our project, we read from the database and then analyze it efficiently and try to obtain scientific results.
- How do you manage large databases without overloading them?
Now, it has two dimensions. First, you want the code you wrote to run for a certain period of time, which means speed, and secondly, since the amount of data you run is large, it does not fit in the memory, first of all, we need large disks, then random access, that is ram, when processing the data. Now, for the code to work well, you need good algorithms, so it needs to be relatively simple without doing too many unnecessary operations and proportionally to the volume you're trying to extract. Apart from this, you need to be careful while storing the data at work. That's why phyton is doing it in the background, and you have to be careful at some points.
- How do you ensure the protection of confidential information?
In fact, we do not have any confidential information, in other words, since the thesis is made with the citizen tax, all data must be open to the public with the decision of the senate. Before going public, some tests are done on the data. There is no such thing as data privacy in our field.
- How has the pandemic affected your research process?
Actually, it didn't affect me much because I work numerically. The telescopes were turned off, so the observers were somewhat affected, and the amount of data flowing into the field decreased. As a matter of fact, we can say that it has been beneficial. I can say that the pace of academic studies has increased in terms of articles. People may have started to produce more articles when they couldn't stay at home and find something to do.
- What are the management dimensions to be considered in the implementation of the research project?
While managing a research project, certain people work under you, whether they are doctoral students, post-docs, and sometimes technicians in order to manage them well, there must be a certain time phenomenon. All these must be pre-planned to get the allowance you have to present them already. A PI, that is, principal investigator, needs to be an expert. In order to do a particular project, how many doctoral students, how many postdocs should they study, how they should work, in which environments they should work, what kind of resources (computers etc.) they need to be followed well.
- Do you use a Gantt Chart and is it useful for management?
Of course, each project has such a gantt chart. When requesting a project from NASA, these graphics are usually used. I am not yet in PI status, but it usually happens in the projects I am involved in and we have to look at that chart over and over to see if we are on the right track.
- What can be done with conflicting results in a research project?
Contradiction actually means the potential for discovery. Every discovery comes with a certain contradiction. My perception is a conflict over model and data. For example, you take a data and pass it to the model, and if these two do not match, either the data is incorrect, the margin of error of the data is incorrect, or the model is incorrect, or the margin of error of the model is incorrect. It has to be one of the four, so discovery occurs.
For further information; https://www.tansudaylan.com/
