Deepnote Reviews & Product Details

Having used Google Colab for many years, Deepnote kicks things up a notch in terms of capabilities. It's unlocked greater insights across the team by having persistent storage, easier external integrations and fantastic visualisation capabilities, allowing us to present data more professionally than before. It's also closer to the experience of using Google Docs where multiple team mates can edit in real time. Easy to use and scales well from junior to senior needs. Review collected by and hosted on G2.com.

It does what it does great, though I'll still use Colab for my own personal work as I find it easier to get going quickly, whereas Deepnotes strengths lie with collaboration across a team. Review collected by and hosted on G2.com.

Easy to use interface, similarity to Google colab makes it easier for developers to build scripts and models.

Also it's integrations with google products, and many datawarehouses, makes it easier to connect and perform ML related tasks on our own data directly from warehouse (Snowflake for us) Review collected by and hosted on G2.com.

Deepnote’s AI is useful when working with individual blocks, but it often falls short in terms of "agent-like" project awareness. It can have difficulty maintaining context throughout an entire notebook, occasionally overlooking explicit instructions or losing track of variables defined in cells that are farther apart. This is especially noticeable when compared to more advanced AI-native IDEs like Cursor. Review collected by and hosted on G2.com.

I think the AI is, in general, very effective. It creates graphs very well. I like the easy integration of Python and SQL. I am not experienced in Python but I am in SQL. It's nice to be able to move back and forth and use the AI to create good visualizations. Review collected by and hosted on G2.com.

Sometimes the AI agent can go a bit crazy and make a bunch of relatively useless blocks of code. Also, it is really annoying that I can't expand the menu bar to see project names. Sometimes we have similarly named projects that I can't see the full name of. Review collected by and hosted on G2.com.

I appreciate that it enables analysts and data scientists to concentrate on their analysis rather than spending time managing their collaborative development environment. Review collected by and hosted on G2.com.

The only drawback I’ve noticed so far is that, to my knowledge, there isn’t currently a Claude Code integration available. I think everyone would appreciate this! Review collected by and hosted on G2.com.

I like the ease of use and the “documentation” generated by the notebooks. I mainly consume dashboards created by the product team, but I can quickly verify how they arrived at the numbers and the conclusions they’re presenting. Review collected by and hosted on G2.com.

I would like to be able to interact with the code in a editor of my choice. Review collected by and hosted on G2.com.

It’s helping improve the team’s capabilities in Data Analytics and enabling the Data team to experiment and build data apps. Review collected by and hosted on G2.com.

AI Assist is still quite limited. It would be much more useful if it included additional features that support building AI apps, or if it made it easier to integrate AI capabilities into more traditional data apps.

The dashboard and data visualization features also need improvement to provide a better end-user experience, especially in terms of usability and overall polish. Review collected by and hosted on G2.com.

Deepnote's ability to have multiple people working on a single document at the same time is a must-have for data science or analytics teams. It's like a Google Docs that came to Jupyter Notebook. This makes it easy to work across teams and streamline the workflow with little friction, despite daily use. Deepnote's AI integration also means that we can focus on just writing the code and leave it up to the AI to automate the syntax hygiene and documentation. Review collected by and hosted on G2.com.

The integration across different coding languages like queuring a data frame in SQL, then using it in R, or importing in custom or proprietary functions from a corporate GitHub is either missing, difficult to implement, or clunky. In these situations, it would be great to have it more streamlined and have a more dedicated environment running in the background to mimic the UI of other cloud services, like Posit. Review collected by and hosted on G2.com.

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. Review collected by and hosted on G2.com.

Deepnote could do better to create more affordable options for teams. The price for example means only a few members in our organisation have access e.g. I can't share my notebook with people outside the data team as they aren't subscribed. Review collected by and hosted on G2.com.

The power in the query processing. It's great when doing heavy queries and data visualizations - all in one. Review collected by and hosted on G2.com.

You should create a feature on downloading .csv files or .xlsx each query or data results. That would be helpful to later export it to sheets. Review collected by and hosted on G2.com.

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. Review collected by and hosted on G2.com.

The AI agent is helpful, but I still prefer using Claude Code directly. When prompting the AI agent in Deepnote, it creates many additional cells, which makes for a jarring experience—I often have to go back and clean them up. My workflow currently consists mainly of editing scripts in Claude Code and then copying them back into Deepnote. For data analysis, however, I do find the AI agent helpful for generating quick visualization scripts. Review collected by and hosted on G2.com.