There are problems with hiring, retaining and developing
Posted: Wed Jan 22, 2025 8:53 am
[ 20 ,-2,+4]
Reksoft Consulting presented a study on the problems faced by Russian Data specialists when developing and implementing AI-based solutions. The authors conducted the study using in-depth interviews: they talked to Data Science team leaders, Data Science specialists, Data Science department heads, Chief Data Officers (CDOs), and Technical Directors. The interview participants work in IT, heavy industry, medicine, retail, and finance.
The authors divided the identified problems into five areas. The first covers the difficulties in the interaction between business and data specialists. The main problem is that companies do not understand the capabilities and principles of AI, implement it either "for show" or place too much hope on the technology. Also, businesses do not understand what economic effect AI-based solutions bring and how to evaluate it. Data specialists do not understand the goals of the business. Because of this, it happens that companies invest in AI, but do not see the result or have difficulties with the adoption of solutions.
The second set of challenges is data-related. Not all companies have sufficiently automated business processes or a mature data infrastructure, and existing data collection methods are not adapted to interact with artificial intelligence.
There are barriers related to development and technology management. Development processes and the technology stack change rapidly, so businesses must be flexible to adapt to circumstances. At the same time, the study participants said that everything is often complicated by the lack of established standards for AI development and approaches to working with external solution developers.
The authors also highlighted the difficulties with the transfer into operation and support of AI solutions. Difficulties arise due to the disorganization of the transfer process kenya whatsapp number database and the lack of clear criteria for acceptance of decisions, including in terms of economic effect. In addition, there is no structured process for monitoring, supporting and developing AI solutions. Approaches to information security are not adapted for the implementation of AI solutions and assessment of their risks.
Data specialists. HR specialists do not know how to work with such employees.
At the same time, the situation in young and mature companies differs. "For a specific company, the priority of problems depends on the level of maturity: new companies talk more about problems of interaction with business, as well as with the quality and availability of data, and leading companies are more concerned about how to properly build an operational model for the successful scaling of AI solutions - development and technology management, transfer into operation and support," Alexey Bogomolov, director of the Transformation Strategy practice at Reksoft Consulting (a division of the Reksoft technology group that deals with transformational and strategic consulting), told ComNews.
Maxim Zharov, key account manager at Sinimex-Informatika JSC (Sinimex is a developer of IT systems for business), told ComNews that it is difficult for young companies to break into the niche, since the entry barrier has suddenly become high due to the popularity of generative networks and the surge in interest in ready-made AI tools. "Large IT companies want to optimize their developments and the process of their monetization, since the "teething problems" of the projects have already been overcome," says Maxim Zharov.
Read also
Generative AI has failed to live up to business expectations
The growth in labor productivity when using generative AI, as demonstrated by the practice of the first attempts to use it to create advertising and marketing materials in Russia over the past year, turned ou
Reksoft Consulting presented a study on the problems faced by Russian Data specialists when developing and implementing AI-based solutions. The authors conducted the study using in-depth interviews: they talked to Data Science team leaders, Data Science specialists, Data Science department heads, Chief Data Officers (CDOs), and Technical Directors. The interview participants work in IT, heavy industry, medicine, retail, and finance.
The authors divided the identified problems into five areas. The first covers the difficulties in the interaction between business and data specialists. The main problem is that companies do not understand the capabilities and principles of AI, implement it either "for show" or place too much hope on the technology. Also, businesses do not understand what economic effect AI-based solutions bring and how to evaluate it. Data specialists do not understand the goals of the business. Because of this, it happens that companies invest in AI, but do not see the result or have difficulties with the adoption of solutions.
The second set of challenges is data-related. Not all companies have sufficiently automated business processes or a mature data infrastructure, and existing data collection methods are not adapted to interact with artificial intelligence.
There are barriers related to development and technology management. Development processes and the technology stack change rapidly, so businesses must be flexible to adapt to circumstances. At the same time, the study participants said that everything is often complicated by the lack of established standards for AI development and approaches to working with external solution developers.
The authors also highlighted the difficulties with the transfer into operation and support of AI solutions. Difficulties arise due to the disorganization of the transfer process kenya whatsapp number database and the lack of clear criteria for acceptance of decisions, including in terms of economic effect. In addition, there is no structured process for monitoring, supporting and developing AI solutions. Approaches to information security are not adapted for the implementation of AI solutions and assessment of their risks.
Data specialists. HR specialists do not know how to work with such employees.
At the same time, the situation in young and mature companies differs. "For a specific company, the priority of problems depends on the level of maturity: new companies talk more about problems of interaction with business, as well as with the quality and availability of data, and leading companies are more concerned about how to properly build an operational model for the successful scaling of AI solutions - development and technology management, transfer into operation and support," Alexey Bogomolov, director of the Transformation Strategy practice at Reksoft Consulting (a division of the Reksoft technology group that deals with transformational and strategic consulting), told ComNews.
Maxim Zharov, key account manager at Sinimex-Informatika JSC (Sinimex is a developer of IT systems for business), told ComNews that it is difficult for young companies to break into the niche, since the entry barrier has suddenly become high due to the popularity of generative networks and the surge in interest in ready-made AI tools. "Large IT companies want to optimize their developments and the process of their monetization, since the "teething problems" of the projects have already been overcome," says Maxim Zharov.
Read also
Generative AI has failed to live up to business expectations
The growth in labor productivity when using generative AI, as demonstrated by the practice of the first attempts to use it to create advertising and marketing materials in Russia over the past year, turned ou