As a Data Scientist in the banking sector, I strongly believe that the adoption of Data Science and Machine Learning could transform older, traditional banks into more digitally savvy banks capable of competing with the rise of more digitally-driven ones of the modern age. AI adoption can benefit other industries as well. The findings from the latest McKinsey Global Survey about the state of AI in 2021 indicate that AI adoption continues to grow and that the benefits remain significant. A majority of McKinsey survey respondents now say their organizations have adopted AI capabilities, as AI’s impact on the bottom line is growing.
However, operationalizing and scaling machine learning to drive business value can be challenging. My experience has shown that, while many businesses have started diving into it, only a few data science projects actually make it to production. Moving from the experiment phase of ML to real-world deployment is difficult, as the journey requires finetuning ML models to fit the practical needs of a business and ensuring the solution can be implemented at scale.
Models as part of an experiment are good, but models in production are great. MLOps, as the name implies, brings operationalization to the table, providing resources for bringing models from test environments into production.
The goal of this notebook is to extract insights from the responses of 2022 Kaggle Machine Learning & Data Science Survey about the state of AI Adoption and ML Operationalization in the industry in 2022 as well as about the Data Science landscape in the market. As I'm curious to see how the MLOps and AI adoption progressing in other organizations and what's the current trends in Data Science I'll try to enlighten the following main topics:
-
What's the state of Machine Learning adoption in the enterprise today?
- What's the percentage of enterprises deploying data science and machine learning in production today?
- Does the company's size or sector play a role in AI Adoption? Are larger companies more likely than smaller companies to have deployed AI in their organization?
-
What's the enterprise AI tech stack?
- Are Cloud-native solutions a must-have for business today?
- What are the most popular tools for Data Storage, Data Management, AutoML, Business Intelligence, etc.?
- What frameworks and libraries are commonly used in the market for Machine Learning and Data Science?
- Are transfer learning methods mature enough to be used in the business environment?
- Do we really work with big data and deep learning methods to such an extent that we need specialized hardware for ML models training?
-
AI Careers & Job Outlook in 2022:
- What are the top AI job positions?
- What does an AI professional do?
- What are the professional AI skills in demand for 2022?
-
AI Salary Overview
In order to have as much as I can a representative dataset for the analysis, I'll keep in the dataset only the professionals, namely the respondents that fulfill the criteria listed below:
- currently are not students (answer No the Q5 question)
- currently are employed (They didn't answer "Currently not employed" to the Q23 question)
- have answered in what industry they are currently employed (or their most recent employer if retired) - Q24 question has an answer, not None
As it can be seen below, ~ 37.9% of the total responses meet the above criteria and the analysis will be conducted based on these responses.