2025 Python Survey Unveils AI Surge, Web Revival, and Performance Boosts

Welcome to the highlights, trends, and actions from the eighth annual Python Developer Survey, a joint effort by the Python Software Foundation and JetBrains PyCharm team.

Michael Kennedy, host of the Talk Python to Me podcast, analyzed more than 30,000 responses to extract the most important insights.

Key Demographics

50% of respondents have less than two years of professional Python experience.

86% use Python as their primary language for writing programs, building applications, and creating APIs.

51% of developers engage in data exploration and processing, mainly with pandas and NumPy.

46% use Python for web development, with FastAPI usage rising from 29% to 38%.

Version Adoption

83% of developers still run Python versions released a year or more ago. Only 15% use the latest release. Reasons for not upgrading include satisfaction with current version (53%) and lack of time (25%).

Upgrading to newer runtimes yields significant performance gains: Python 3.11 users see ~11% speed improvement and 10‑15% lower memory usage; moving from 3.10 or older to 3.13 can boost speed by ~42% and cut memory by 20‑30% without code changes.

Cloud and Cost Implications

Many developers run Python in Docker containers, which simplifies upgrading. For medium‑size enterprises, a 42% speed gain could save $420 k annually; for large enterprises, up to $5.6 M, based on typical AWS EC2 spend.

Rust as a Performance Partner

Rust is increasingly used in Python ecosystems: about 30‑33% of binary extensions now include Rust code. Projects like Polars, Pydantic, and new servers such as Granian illustrate this trend.

Typed Python Momentum

New high‑performance type checkers written in Rust—Astral’s ty and Meta’s pyrefly —are emerging, offering fast language‑server support.

Database Preference

PostgreSQL remains the dominant database for Python developers, rising from 43% to 49% usage.

AI‑Driven Coding Agents

69% of respondents plan to try AI coding agents; 49% intend to adopt them within a year. Tooling such as Junie (JetBrains) and other agents can increase productivity by roughly 30%.

Async and Threading Evolution

Async/await is now core to Python, and Python 3.14 will introduce “free‑threaded” Python, removing the GIL and enabling true parallelism.

Actionable Recommendations

Learn uv for fast package management and virtual environments.

Upgrade to the latest Python version (e.g., 3.13) using uv venv or Docker.

Explore AI coding agents to boost efficiency.

Study basic Rust concepts to understand performance‑critical libraries.

Invest in understanding true threading and concurrency.

Prioritize beginner‑friendly documentation and tutorials.

Visual Data