Databricks Reviews

“its a complete tool”

Pros

In real project work, it makes a huge difference not having to jump across multiple tools for data engineering, analytics and machine learning. The collaboration side is also a standout. The notebooks feel fluid and interactive so teams can work together without things turning chaotic. For serious data projects, it almost feels like Google Docs for large-scale data work. Another thing I appreciate is how well it handles big volumes of data without making the experience feel overly complicated. Even with large datasets, the platform stays approachable and can scale up when needed. From an AI and ML perspective, it also fits very naturally. You can build,

Cons

It seems a bit overwhelming at first. There are clusters, notebooks, jobs and workflows so a lot is happening at once and if you're new to the platform, it takes time to understand how all the pieces connect. Managing cost is another downside. The platform is undeniably powerful but spending can climb fast if cluster usage or auto-scaling settings are not watched carefully. It really takes discipline and regular monitoring to keep costs under control.

“unified data platform”

Pros

This tool makes data management much simpler. I used to rely on multiple tools for different data tasks and that setup was far less convenient, while here everything is connected. The notebook functionality is another major benefit especially when working with PySpark and that is something I really value. It also lets me make updates and changes quickly without needing too much setup beforehand. Collaboration has improved as well since my team can work on their own projects at the same time while still keeping track of overall progress. Version control can feel a little unclear at times, though. Performance wise, this tool has been very efficient in handling big data and generally runs without noticeable delays. Automatic cluster scaling saves both me and my team time on the infrastructure side since we do not have to spend extra effort planning or adjusting resources. There are a few minor UI slowdowns now and then but overall because it is so effective for implementation and integration, it keeps me coming back to it regularly.

Cons

One thing that frustrates me is the UI. After spending more time in the platform, switching between notebooks and clusters can start to feel irritating. Cost management is another issue because expenses can pile up quickly if you are not paying close attention. Clusters sometimes keep running longer than needed without me or my team noticing which drives project costs up unnecessarily. Debugging can also be quite complicated especially when dealing with more complex pipelines since it often takes extra time and effort to figure out exactly where something went wrong. Customer support has also been inconsistent at times which can lead us into situations that are more frustrating than helpful.

“Much faster content creation”

Pros

It's helped our team produce content more quickly.

Cons

I noticed that the product review and approval process at our company has been taking a long time.

“Best Spark processing and sharing”

Pros

Working with Databricks for big data processing and data engineering in PySpark has been really effective. It lets me handle terabytes of data without issues thanks to the Spark architecture. One feature I really appreciate is Unity Catalog and its access framework because it makes sharing data across the organization much more manageable while still letting me control permissions like Select, View and other access on Delta tables by role or team. The setup at the beginning went very well and the integration with Microsoft Fabric is something I have also valued.

Cons

The area that could use some improvement is the billing experience especially since I am using it through Azure.

“all in one data workspace”

Pros

It really stands out by bringing data engineering, analytics and machine learning together in one unified workspace. Shared notebooks make teamwork much more convenient and the integration with big data tools helps cut down on time. It makes complicated workflows easier to manage while still giving me the advanced capabilities I need when the work gets more demanding.

Cons

It does have a few downsides especially when it comes to cost. It can feel pretty expensive for smaller teams or projects and cluster setup along with cost management can be a little tricky to handle. The interface is powerful but it can also feel overwhelming if you're new to it and troubleshooting distributed jobs is not always as simple as I would prefer.

“Its a unified data platform”

Pros

Databricks does a really good job of simplifying big data work by bringing data engineering, analytics and machine learning into one unified platform. The tight Spark integration and strong scalability make working with large datasets much more efficient.

Cons

Pricing can climb pretty quickly with heavy usage especially when clusters aren't optimized well. On top of that debugging and monitoring jobs can feel a bit less user friendly than with more traditional tools.

“effective lakehouse for analytics”

Pros

It is a powerful platform that combines data engineering, AI/ML and SQL analytics in one collaborative workspace.

Cons

This tool is expensive which can add up quickly with frequent cluster usage and workloads that aren't well optimized.

“Powerful workspace and cluster control”

Pros

Databricks stood out for how manageable it is. The cluster management, unified workspace, performance optimization and versioning have all been extremely valuable in day to day use. Having all the tools available in one console is really convenient especially for large-scale data engineering work. The initial setup was also very easy which made getting started with the platform pretty pleasant.

Cons

There really isn't much to complain about.

“Unified workspace boosts team speed”

Pros

Having data engineering, analytics and machine learning in a single, well-organized workspace is the biggest advantage of this tool. It cuts down on wasted time, makes collaboration much simpler and helps teams work faster when dealing with large volumes of data.

Cons

One part that gets frustrating with it is Auto Loader when source data keeps changing often especially if column names or data types shift unexpectedly. For instance a field such as customer_id might suddenly appear as cust_id or a column that used to be a string may start coming through as an integer, creating schema drift and disrupting downstream processing. It is also inconvenient when schema inference is not completely accurate, particularly with nested JSON or other semi-structured data because that leads to extra manual fixes and ongoing maintenance just to keep pipelines running properly.

“An all in one data powerhouse”

Pros

This tool is amazing as having it means we can do data processing, analytics and pipeline work without jumping between many separate tools. It works very well with large amounts of data and not having to manually manage clusters saves a lot of time. The platform also handles collaboration and experimentation well so testing new ideas is easy and practical.

Cons

The part that still needs improvement is cost management. If cluster usage is not watched closely, expenses can climb quicker than expected. It would be much better to have more detailed visibility into where costs are coming from. Built-in alerts or recommendations when spending starts rising unexpectedly would also make a big difference.

“we rely on it for a lot”

Pros

Databricks sits at the center of our day to day data work and we rely on it to build and run pipelines across a full medallion architecture, starting with data extraction from SAP and Arkieva through to datasets that are ready for reporting. The notebook environment is also user friendly, the platform offers a huge range of capabilities and Asset Bundles have really strengthened our CI/CD process with Azure DevOps. Its cloud integrations came together without much friction and once everything is configured, it tends to run reliably. There is definitely a learning curve for newer teammates especially when they start dealing with Unity Catalog and DABs and costs can rise if cluster settings are not watched closely. Even so support has been decent and the documentation is good enough that we rarely need to submit tickets. All in all it is a very capable platform that covers a lot in one place and it would be hard to picture our data engineering workflow without it.

Cons

Costs can climb pretty quickly if cluster sizing and job configurations are not managed carefully so keeping the environment optimized takes ongoing attention. On top of that newer team members usually need time to get comfortable, particularly with Asset Bundles, Unity Catalog and setting up the CI/CD components correctly.

“productivity boost !!”

Pros

It completely changed the way our team manages end to end data workflows. Several things really stand out. The notebook UI is user friendly and the SQL editor is polished so moving between Python, SQL and Scala in the same workspace cuts down a lot of context switching. On the integration side the native connections to Azure, Unity Catalog and Delta Sharing save us from spending so much time on setup and plumbing. Lakehouse Federation was also a nice surprise since we can query external sources without having to move the data first. Performance has been impressive too. Delta Lake's auto-optimization and liquid clustering made a noticeable difference in our query speeds and Photon has been a huge win for heavy aggregations and near real-time dashboards. The DBU pricing model takes a little time to understand but once we consolidated our data warehouse, ETL and ML tools into it, our overall infrastructure costs dropped quite a bit. Onboarding has also been faster thanks to Databricks Academy and the built-in documentation and the community forum has been genuinely helpful for more niche questions. The AI features are another plus as Genie helps business users ask questions in plain English and get reliable answers which has clearly reduced ad hoc requests to our data team, while Databricks Assistant in notebooks speeds up both code generation and debugging.

Cons

Nothing to share