Databricks Data Intelligence Platform
Databricks, Inc.External reviews
763 reviews
from
and
External reviews are not included in the AWS star rating for the product.
BI and Data Engineering in One Place, with AI-Assistant
What do you like best about the product?
Possibility to combine data warehousing and data lakes into a “lakehouse.” So I can do BI and data engineering all in one place instead of stitching together multiple systems.
Using AI to improve and make faster the SQL writing and execution
Using AI to improve and make faster the SQL writing and execution
What do you dislike about the product?
Unity Catalog is powerful, but setting up fine-grained access control across data, schemas, and workspaces can become tricky, especially in larger organizations. The UX/UI of some parts of the platform feels polished, others less so.
What problems is the product solving and how is that benefiting you?
Databricks is essentially solving fragmentation and inefficiency across the data lifecycle and the benefits come from removing a lot of friction between teams, tools, systems and data.
Unified Platform with Powerful Features, Needs Faster Cluster Startups
What do you like best about the product?
I appreciate how Databricks brought everything onto one unified platform, allowing our teams to collaborate in shared notebooks and ensuring data consistency with Delta Lake's ACID transactions. My favorite feature is Auto Loader, which automatically ingests new data files as they land in cloud storage, saving our team 2-3 hours a week on manual pipeline monitoring. Unity Catalog has been a game changer for us, providing a central place for governance and access control, which before was a mess. The initial setup was straightforward, and we had our first cluster and notebooks connected to S3 within a day, which was impressive given the platform's power. The workspace configuration and cloud integration guides are solid to follow.
What do you dislike about the product?
The cluster startup time is something that still catches us off guard. Cold start can take anywhere from 3-5 minutes, which gets frustrating when you are in the middle of an iterative debugging session and just need to test a quick fix. The cost management also needs some upgrades as currently the billing dashboards are improving but it still takes some digging to pinpoint exactly which job or user is driving up spend.
What problems is the product solving and how is that benefiting you?
I use Databricks to unify our data processing and machine learning, reducing pipeline delivery delays by 40%. It enables team collaboration with consistent data, saving hours with the autoloader, and simplifies governance with Unity Catalog.
Streamlined Data Processing with Unmatched Speed
What do you like best about the product?
I use Databricks for real-time data ingestion and processing as well as batch processing. I find it easy to use with PySpark, and I appreciate that it serves as a single platform for both real-time and batch processing. The in-memory processing drastically reduces processing time, and working with dataframes makes handling structured data straightforward. I like the fast execution and the ability to clean, massage, and manipulate data all on the same platform. It's also easy to deploy, and I enjoy the smooth CI pipeline with just one click. The initial setup was quite easy, and the product support made it a cakewalk.
What do you dislike about the product?
Databricks should come up with agentic framework integrated, making it a single stop for Data and AI.
What problems is the product solving and how is that benefiting you?
Databricks offers an easy-to-use platform for both realtime and batch processing. It integrates easily with PySpark and supports in-memory processing, significantly reducing processing time. Dataframes make handling structured data simpler.
Great UI and a Straightforward, Linear Learning Curve.
What do you like best about the product?
The UI is great compared to other providers. It’s easy to work with, and the learning curve feels linear and straightforward.
What do you dislike about the product?
Consumption-based costs are on the higher side, and it can be difficult for users who aren’t proficient in Python or Spark.
What problems is the product solving and how is that benefiting you?
A centralised data warehouse, with notebooks running on top of it for further analysis and ML use cases.
Unified Platform with Scalability and ML Power for Big Data
What do you like best about the product?
I like Databricks for its unified platform, which brings data engineering, analytics, and machine learning together. It simplifies workflow scaling and is easy for handling big data. The collaboration across the team is much smoother, which I really appreciate.
What do you dislike about the product?
I would say cost transparency maybe. User-based pricing can be hard to predict. So the initial setup and cluster configuration can feel complex. Better documentation for that and UI could be more intuitive in some areas.
What problems is the product solving and how is that benefiting you?
I use Databricks to sort ETL pipelines, handle large-scale data efficiently, reduce data processing time, and eliminate data silos. The unified platform improves collaboration between data engineers and scientists, simplifying workflows and making big data management smoother.
Seamless Integration and Scalable Performance with Room for UI Improvement
What do you like best about the product?
I use Databricks to build ETL pipelines and process large-scale data with Spark. I like Databricks most for its seamless integration with Apache Spark, collaborative notebooks, and its ability to handle large-scale data processing efficiently in a unified platform. The seamless Apache Spark integration lets me process huge datasets quickly without worrying about cluster setup, while collaborative notebooks make it easy to work with my team in real-time. The scalable architecture ensures reliable performance even with heavy data workloads. The initial setup of Databricks was fairly straightforward, especially with cloud integration.
What do you dislike about the product?
The UI can feel a bit cluttered at times, cluster startup times can be slow, and the pricing can get expensive for smaller projects or prolonged usage.
What problems is the product solving and how is that benefiting you?
I use Databricks to efficiently process large-scale data, simplify ETL workflows, and collaborate with my team in a unified environment, gaining faster data-driven insights.
Databricks Unifies Engineering and Analytics for Scalable Spark Pipelines
What do you like best about the product?
What I like best about Databricks is that it brings data engineering, processing, and analytics into one platform.
From my perspective, it makes it much easier to build and manage scalable pipelines with Spark without worrying too much about infrastructure.
From my perspective, it makes it much easier to build and manage scalable pipelines with Spark without worrying too much about infrastructure.
What do you dislike about the product?
What I dislike about Databricks is that cost control can get tricky if clusters are not managed properly.
Also, debugging distributed jobs is not always straightforward, and sometimes the UI feels a bit heavy when you just want quick insights
Also, debugging distributed jobs is not always straightforward, and sometimes the UI feels a bit heavy when you just want quick insights
What problems is the product solving and how is that benefiting you?
Databricks solves the problem of handling large scale data processing and fragmented tools.
For me, it brings ETL, streaming, and analytics into one place, which reduces pipeline complexity and speeds up development and troubleshooting.
For me, it brings ETL, streaming, and analytics into one place, which reduces pipeline complexity and speeds up development and troubleshooting.
Powerful Unified Analytics with Seamless Governance and Effortless Scaling
What do you like best about the product?
What I like best about Databricks is its powerful and unified analytics ecosystem. Features like Unity Catalog and Metastore make data governance and access control seamless, while the Lakehouse architecture combines the best of data lakes and warehouses. PySpark support, dbutils, and collaborative workspaces make development efficient, and serverless compute simplifies scaling without infrastructure overhead.
What do you dislike about the product?
What I dislike about Databricks is the slow startup time of all-purpose clusters, which can interrupt workflow and reduce productivity. Additionally, Git integration can feel a bit sluggish at times, especially during commits or syncing, making version control less seamless than expected.
What problems is the product solving and how is that benefiting you?
Databricks solves the challenge of managing end-to-end data workflows by providing a unified platform for data engineering, data science, and analytics. It allows seamless data processing, transformation, and model development within a single environment.
This benefits me by simplifying my workflow as both a data engineer and data scientist, reducing the need to switch between tools. Additionally, its integration with Azure Data Factory enables smooth job orchestration and triggering for higher environments, making deployments more efficient and reliable.
This benefits me by simplifying my workflow as both a data engineer and data scientist, reducing the need to switch between tools. Additionally, its integration with Azure Data Factory enables smooth job orchestration and triggering for higher environments, making deployments more efficient and reliable.
Unified Data Platform, Minor Cost and Complexity Challenges
What do you like best about the product?
I like that Databricks provides a unified platform for data engineering and data science, eliminating friction across teams and enhancing the ability to accelerate development and deployments. It works especially well for end-to-end CICD pipelines.
What do you dislike about the product?
Well, in terms of what can be improved, I think, perhaps the cost management. If this can be looked into to make it more cost efficient for users, it will go a long way. And in addition to that, operational complexity sometimes presents a complex platform for new users to navigate easily. So if this can be addressed, then I think it should be a lot easier for engineers to work with.
What problems is the product solving and how is that benefiting you?
I use Databricks for scalable workflows across multi-cloud environments, solving data silo unification and minimizing bottlenecks in complex data processing. It optimizes cost and governance while providing a collaborative workspace, real time data ingestion, and enhanced system reliability and performance.
Unified Data Workflows with Databricks
What do you like best about the product?
I really like Databricks for its collaborative lake house environment, which has been key in unifying our data engineering and machine learning workflows. It bridges the gap between our engineering and analytics teams, allowing us to run BI and AI on a single platform. Additionally, the initial setup was surprisingly fast from a workspace perspective, especially with the native integration in Azure.
What do you dislike about the product?
The learning curve is quite steep for non-engineers. We've also had to be very diligent with cost monitoring as auto scaling clusters quickly lead to unexpected expenses if not managed strictly.
What problems is the product solving and how is that benefiting you?
Databricks solved our data stack fragmentation by unifying storage lakes and warehouses. It bridged the gap between engineering and analytics, letting us run BI and AI on a single platform.
showing 1 - 10