Skip to content
#Data

Code of Leadership 50 - Interview with Sergey Mikhalev about Odnloklassniki, VK, T-Bank and Data Platform (Data column)

#Data #Software #Engineering #Databases #Architecture #Devops #Management #Career #DistributedSystems #SystemDesign #Leadership

Another one. podcast An interesting guest came to me, Sergey Mikhalev, who works as the technical director of the data platform at T-Bank. Before that, Sergey managed to put a hand to the date platforms of Odnoklassniki and VKontakte. In this video, Sergey and I discussed his career path, which was connected with big data all the way. At the same time, we discussed how this concept has changed over time and what the state of the art looks like now, as well as why we are moving from T-Bank to Greenplum. (mpp database) a more fashionable and youthful approach with DLH (data lake house) And how does the federated approach with data mesh work, and how does the a/b platform work on top of that? Or, to be more precise, the list of things we have discussed 2 hourly

Introduction and introduction to the guest Early years, education and professional choice Starting a career, optimizing databases, migrating data and growing to a team leader

  • Building competencies
  • Work in Odnoklassniki on the date infrastructure Engineering and experimental culture of “Classmates” Importance of data and lots of a/b experiments as a sign of mature product Development of the data platform in Odnoklassniki and transition to the position of technical director of the VK data platform Transition to T-bank and discussion of the data culture adopted in T
  • Level of engineering culture, duty, log processing, AB tests, interfaces for experiments. Greenplum architecture, parallelism, constraints, cluster expansion
  • Reliability, switching risks, backup system. Greenplum Limitations New data platform architecture with separation of compute and storage A new paradigm of working with data: the introduction of the date of contracts Federalization of engineering data and transfer of responsibility to product teams Services and data products, pipelines, interaction in the ecosystem Formats of work: office vs. remote
  • Work-life balance The future of IT and the potential of biomedicine Self-development and motivational support
  • Completion and thanks

The podcast release is available in Youtube, VK Video, Podster.fm, Ya Music.

#Software #Engineering #Databases #Architecture #Devops #Management #Career #Architecture #DistributedSystems #SystemDesign #Leadership