Explore data with R, create plots, and explain the results beside your code. Deepnote gives you a browser-based workspace for keeping an analysis together and sharing it with your team.
Start with a dataset or bring an existing notebook. Work through your calculations, inspect the outputs, and give others the context they need to understand your findings. No local R installation is required.
summary_2007 <- gapminder %>% filter(year == 2007) %>% group_by(continent) %>% summarise(median_life = median(lifeExp)) summary_2007
continentchr Africa20% Americas20% 3 others60% | median_lifedbl 52.9 – 80.7 | |
|---|---|---|
| 1 | Africa | 52.93 |
| 2 | Americas | 72.90 |
| 3 | Asia | 72.40 |
| 4 | Europe | 78.61 |
| 5 | Oceania | 80.72 |
Keep the source data, transformations, plots and interpretation together. R notebooks let you develop an analysis in visible steps, so another person can follow how you reached a conclusion.
Explore a distribution, compare groups, or investigate a pattern over time. Add an explanation where a choice affects the result, such as excluding missing values or changing the population you analyze.
Select an available R environment when setting up your project. Check the runtime and required libraries before you begin, and add the data your code uses. Your analysis runs in the hosted environment while you work through your browser.
install.packages("gapminder") library(gapminder) library(ggplot2)
ggplot(gapminder, aes(gdpPercap, lifeExp)) + geom_point(alpha = 0.4) + scale_x_log10() + theme_minimal()
Load installed packages with library() and add missing packages with install.packages(). Packages that depend on system libraries may need additional environment configuration.
Use libraries such as ggplot2 to create plots alongside the calculations they illustrate. Keep the transformation code visible so readers can understand what the chart represents.
Record the data source, runtime and package versions, along with the assumptions behind your calculations. Run the notebook from a clean environment to check that it works in the intended order.
For analyses that need specific dependencies, configure a custom environment. Project history helps you revisit earlier work; reproducible results also depend on the data and environment used to run it.
Deepnote is recognized as a Leader and a Mid-Market High Performer in G2’s Data Science and Machine Learning Platforms category.
“Deepnote makes collaboration on data science and research projects effortless. I really like its real-time editing, version control, and easy cloud-based setup. It’s great for teaching, code reviews, and sharing results without complex configuration.”
Read the full Deepnote review“Moving to Deepnote has transformed how we collaborate and share analysis, research, and experiments at PredictAP. The integrations allow us to connect to different data sources and easily run analysis with the help of AI-generated scripts. Notebooks are easily accessible, making experiments more reproducible without the overhead of setting up developer environments.”
Read the full Deepnote review“I love that Deepnote enables me and my team to share work and collaborate, thus building off of each other's strengths. I have loved the AI automations on Deepnote that enable me to ask a question in natural language and for Deepnote to generate working code for me to run out of the box.”
Read the full Deepnote reviewFollow an experiment from raw data to a shared results app. Connect your data, analyze it with R and AI, and help your team understand what changed.
| participant_id | experiment_group | date | converted | |
|---|---|---|---|---|
| 1 | 1042 | Control | 2026-10-01 | 0 |
| 2 | 1043 | Variant B | 2026-10-01 | 1 |
| 3 | 1044 | Variant B | 2026-10-02 | 0 |
| 4 | 1045 | Control | 2026-10-02 | 1 |
| 5 | 1046 | Variant B | 2026-10-03 | 1 |
See how our community members use R notebooks to investigate data, from changes over time to regional patterns.
Deepnote Agent Workspace combines notebooks, data connections,
and business knowledge so agents can work with your data’s context.
Connect through Cursor, Claude Code, Codex, and other MCP-compatible tools. Access Deepnote notebooks and data within existing permissions.
Schedule notebook workflows or trigger them through the API. Automate reports, data checks, and recurring analyses.
Keep metric definitions, assumptions, analysis steps, and business context alongside your code. Turn trusted work into reusable skills, agents, and data apps.
Find the setup and sharing instructions for your next analysis:
Bring your data, and keep your code, plots and interpretation together. Build an analysis that your team and agents can reuse.
