The results of the annual survey have been published by the Python Software Foundation (PSF) and software maker JetBrains. More than 25,000 developers from around the world participated in the “Python Developers Survey 2023”. The seventh edition of the survey shows the most common use of the programming language, as well as the most popular libraries and frameworks. Half of those surveyed use Python professionally and privately, 28 percent mainly for personal purposes or side projects and 21 percent mainly for work.
Advertisement
Data processing with Pandas and NumPy, visualization with Plotly Dash
Python is most often used for web development – 21 percent of study participants mentioned this purpose. Machine learning and data analysis come in second and third with ten percent each. More unusual uses include mobile or embedded development, each of which makes it into one percent.
Python Developers Survey 2023: The most common uses of Python are web development, machine learning, and data analysis.
(Image: Python Software Foundation/JetBrains)
The list of Python’s use as a main and second language – multiple entries were possible – was expanded this year to include some selection options. New categories of data engineering and academic research enter fourth and fifth place, respectively, pushing “DevOps / System Administration / Writing Automation Scripts” to sixth place. As in last year, data analysis (44%) is in the top three, followed by web development (44%) and machine learning (34%).
Data analysis tops the top 10 areas of application for Python as a main and second language (multiple answers possible).
(Image: Python Software Foundation/JetBrains)
last year The following picture emerged regarding the purposes of Python:
Top areas of application for Python as a main and second language in the “Python Developers Survey 2022” (multiple answers possible).
(Image: Python Software Foundation/JetBrains)
Almost half of the Python developers surveyed are involved in data exploration and processing and use various tools for this. The libraries Pandas with 77 percent and NumPy with 72 percent are primarily used. The cluster computing framework Apache Spark (16%), the open source workflow management platform Apache Airflow (14%) and the DataFrame interface Polder (10%) are far behind. A quarter of those surveyed also create dashboards. Of these, about a third use the Plotly Dash framework, which is based on the JavaScript libraries Plotly.js and React, as well as the web application framework Flask.
Data2Day 2024, on September 18 and 19There are conferences for data scientists, data engineers, and data teams. They focus on different things Data Architecture How else Data Mesh & Co. Prove yourself in practice. In several lectures, experts will examine the question of which new approaches will prevail Large Language Models (LLM), Generative AI and Knowledge Graphs For the work of data scientists.
Machine learning: Scikit-learn before PyTorch and TensorFlow
In the machine learning area, the Scikit-learn library (67%) as well as the PyTorch (60%) and TensorFlow (48%) frameworks are particularly used, while when it comes to training platforms, more than half of those surveyed use Jupyter Notebooks. Amazon SageMaker, Cloud VMs with SSH, Azure ML, and Databricks follow behind in the low double-digit or single-digit percentage range.
Demographics: 21 to 39 years of age and working full-time in the IT sector
But who are the Python developers surveyed? About a third of each are 21 to 29 or 30 to 39 years old. The lower and upper echelons are eight percent who are 18 to 20 years old and three percent who are 60 years old or older. When asked the optional gender question, most respondents said they were male (87%), while female (8%) and “non-binary/genderqueer, or gender non-conforming” (1%) were mentioned significantly less frequently. Three percent declined to provide information about their gender. A fifth of those surveyed come from the United States. India is in second place with nine percent and Germany is third with six percent.
People were also asked about their working life: 62 percent work full-time for a company, twelve percent are studying and six percent are self-employed. The most frequently mentioned industry is, unsurprisingly, IT/software development (38%) and most respondents see themselves in the role of developer/programmer (62%). Rarer mentions include, for example, academic researcher (9%), technical support (8%) and product manager (5%). The most frequently mentioned company size is from 51 to 500 employees, the most common team size is two to seven people. Only three percent of those surveyed work in teams with more than 40 people – which is probably the highest number.
Both these and other results of the survey are On the JetBrains website Also to see Raw data is available as. The study organizers also explain the methodology In detail,
(May)
