BAKU, Azerbaijan, June 9. The Central Bank of
Azerbaijan (CBA) will switch to the "Data Lakehouse" system by the
end of this year, Head of Data Department, Data Engineering
Division at CBA, Mahir Guluzade, said during the 10th International
Finance and Banking Summit taking place in Baku, Trend reports.


"As the Central Bank, we have projects to transition to the
'Data Lakehouse' system and Data Governance by 2026. Currently,
very serious work is being carried out in this direction," he
explained.


Guluzade touched on the importance of data management for the
success of artificial intelligence (AI) projects in the corporate
environment.


He noted that currently, many artificial intelligence products
and corporate solutions cannot be integrated into a real work
environment (production). According to him, the main challenge here
is the lack of a fully ready data infrastructure for AI-ready
data.


"Many believe that if an organization has a data warehouse or a
data lake, then it's already possible to start implementing
artificial intelligence. However, it is imperative that there is a
special layer that ensures that the data to be used here also has
meaning for AI and is correctly interpreted by the system. Just as
there is a semantic layer in analytical data, the establishment of
a 'semantic context layer' is an essential condition here," the
official said.


Speaking about the current situation in the financial sector,
the head of the department said that although structured data
predominates in this field, unstructured data also exists. Modern
solutions, such as "Data Lakehouse," should be used to collect all
this data in a single center.


Strong Data Governance is a key condition







Emphasizing that the most fundamental element for the successful
application of artificial intelligence is strong data governance,
Guluzade pointed out that this system allows AI to understand the
essence of existing data.


"Data Lineage and Quality data governance refers not only to
general rules, but also to the traceability of the source of data
(lineage) and quality (quality), which is of critical importance
for AI. Ontologies and Knowledge Graphs: One of the biggest
challenges is the development of special ontologies and the
construction of Knowledge Graphs that can use the 'Graph RAG'
(Retrieval-Augmented Generation) technology," he pointed out.


Goals and Strategy of the CBA


Speaking about the current projects of the CBA in this
direction, Guluzade also announced the bank's digital transition
plans.


"As the Central Bank, we have projects to transition to the
'Data Lakehouse' system and Data Governance by 2026. Currently,
very serious work is being done in this direction. We believe that
even though it is the last year of our strategy, we will fully
achieve the goals set. Building an artificial intelligence layer,
i.e., ML-Ops (Machine Learning Operations) system on top of this
infrastructure is one of our main priorities, and our relevant
strategy is already ready," he added.