Databricks Data Intelligence Platform
Databricks, Inc.External reviews
636 reviews
from
and
External reviews are not included in the AWS star rating for the product.
Best Collaborative platform for data engineer, analyst and scientists
What do you like best about the product?
Easy to use, it provides one under umbrella platfrom where different teams collborate their work together, which is very helpful for development and data sharing.
What do you dislike about the product?
as of now i dont find any issues, but we can improve on unity catalog side.
What problems is the product solving and how is that benefiting you?
We have different pipelines in databricks, we are utlisinf it for getting spark benifts and colloborative developement and data sharing between teams.
Worth the effort
What do you like best about the product?
Databricks excels at unifying data engineering, analytics, and machine learning into one seamless platform. What I like best is how effortlessly it handles massive data volumes while enabling collaborative development through notebooks. The integration with Apache Spark and the ability to run scalable workloads with ML, SQL, and Python side-by-side makes it a powerhouse for data-driven teams. Its governance and Delta Lake architecture also ensure reliability and security across the data pipeline.
What do you dislike about the product?
While Databricks is incredibly powerful, the learning curve can be steep for non-technical users or teams new to distributed computing. The UI, though functional, can sometimes feel a bit clunky compared to more modern data platforms. Additionally, managing costs in a multi-user environment requires careful governance, especially for teams running large-scale compute-heavy jobs.
What problems is the product solving and how is that benefiting you?
Databricks is helping us break down data silos by centralizing data engineering, analytics, and machine learning into a unified environment. It simplifies handling large datasets, automates ETL processes, and enables real-time analytics and AI-driven insights. As a result, we’ve significantly improved our data pipeline efficiency, reduced time to insights, and empowered both data scientists and analysts to collaborate more effectively using a single platform.
Unlocking Scalable Data Insights with Databricks
What do you like best about the product?
Databricks excels in unifying data engineering, analytics, and machine learning in a collaborative, cloud-based environment. Its support for multiple programming languages (Python, SQL, Scala, R) makes it incredibly flexible. The Lakehouse architecture simplifies data management by combining the best of data lakes and data warehouses. The auto-scaling compute clusters, tight integration with tools like MLflow, and powerful notebooks streamline experimentation and production deployment. I also appreciate the frequent product updates and commitment to open-source technologies like Apache Spark and Delta Lake.
What do you dislike about the product?
While powerful, Databricks has a learning curve—especially for non-technical users or those new to Spark-based architectures. Pricing can escalate quickly if not closely monitored, particularly with always-on clusters. The UI, although improving, still feels unintuitive in certain areas (like managing jobs or cluster permissions). Some integrations, especially with on-premise systems, require additional effort or custom workarounds.
What problems is the product solving and how is that benefiting you?
Databricks addresses the fragmentation between data engineering, data science, and analytics by offering a unified platform. Previously, we struggled with maintaining multiple disconnected tools for ETL, machine learning, and BI. Databricks' Lakehouse architecture allows us to manage structured and unstructured data in a single place, simplifying our data pipelines and reducing operational overhead.
It also improves collaboration across teams—data engineers, analysts, and data scientists can work together in shared notebooks with version control and built-in visualizations. With Delta Lake, we now have ACID-compliant data reliability and time-travel capabilities, which help ensure data quality and reproducibility.
As a result, project delivery times have decreased, and our ability to iterate quickly on models and reports has improved significantly—leading to faster business insights and better data-driven decision-making.
It also improves collaboration across teams—data engineers, analysts, and data scientists can work together in shared notebooks with version control and built-in visualizations. With Delta Lake, we now have ACID-compliant data reliability and time-travel capabilities, which help ensure data quality and reproducibility.
As a result, project delivery times have decreased, and our ability to iterate quickly on models and reports has improved significantly—leading to faster business insights and better data-driven decision-making.
Databricks unifies majority of engineering platforms.
What do you like best about the product?
There are various reasons why I like the Databricks Data Intelligence Platform.
We have been using Databricks as a Unified Platform for All of our Data Workloads (pipelines and models) that encompasses data engineering, data science, analytics and agentic AI.
We love the lakehouse architecture that assists in onboarding traditional data warehousing specialists to Databricks in a fast and scalable way.
We did Unity Catalog upgrade last year that streamlined governance and access control across assets; and did Serverless compute this year that decreased cluster starting/waiting time tremendously.
We have been playing around with Agentic AI space now where we build easy to understand prompts to train agents built on top of datasets. This helps in profiling/slicing/analyzing data by anyone (even a non-technical business person).
BI team has been using Databricks to connect with our Power BI dashboards while Engineering team has been using Databricks to connect with Airflow to create/visualize native DAGs.
Last but not the least, Databricks Notebook concept is awesome.
In the same Databricks notebook, one can have code in multiple languages (Python, Scala, SQL etc.) and each can be flipped at runtime.
Built-in collaborative notebooks with support for multiple languages (Python, Scala, SQL, R) and real-time co-authoring make it easier for teams to iterate together quickly.
We have been using Databricks as a Unified Platform for All of our Data Workloads (pipelines and models) that encompasses data engineering, data science, analytics and agentic AI.
We love the lakehouse architecture that assists in onboarding traditional data warehousing specialists to Databricks in a fast and scalable way.
We did Unity Catalog upgrade last year that streamlined governance and access control across assets; and did Serverless compute this year that decreased cluster starting/waiting time tremendously.
We have been playing around with Agentic AI space now where we build easy to understand prompts to train agents built on top of datasets. This helps in profiling/slicing/analyzing data by anyone (even a non-technical business person).
BI team has been using Databricks to connect with our Power BI dashboards while Engineering team has been using Databricks to connect with Airflow to create/visualize native DAGs.
Last but not the least, Databricks Notebook concept is awesome.
In the same Databricks notebook, one can have code in multiple languages (Python, Scala, SQL etc.) and each can be flipped at runtime.
Built-in collaborative notebooks with support for multiple languages (Python, Scala, SQL, R) and real-time co-authoring make it easier for teams to iterate together quickly.
What do you dislike about the product?
-- Unpredictable costs for small teams. We faced it when one of our agents AI workflow was running for a long time on a very small dataset.
-- Compute + storage based costing means optimization is critical. Every QTR, we have to really keep our Databricks forecasts updated as we don’t want a big deviation between forecast vs. actual.
-- While Databricks supports dashboards (with limitations), it can’t replace BI tools like Power BI, Tableau etc.
-- Compute + storage based costing means optimization is critical. Every QTR, we have to really keep our Databricks forecasts updated as we don’t want a big deviation between forecast vs. actual.
-- While Databricks supports dashboards (with limitations), it can’t replace BI tools like Power BI, Tableau etc.
What problems is the product solving and how is that benefiting you?
We have been playing around with Agentic AI space now where we build easy to understand prompts to train agents built on top of datasets. This helps in profiling/slicing/analyzing data by anyone (even a non-technical business person).
Business teams are also using Power BI dashboards (revenue, corporate accounting, HR, treasury and commerce) fed from Databricks data now.
Business teams are also using Power BI dashboards (revenue, corporate accounting, HR, treasury and commerce) fed from Databricks data now.
All under one hood
What do you like best about the product?
I like that everything you need is in the same platform.
What do you dislike about the product?
It is a bit too easy to overspend, need billing alerts in the platform.
What problems is the product solving and how is that benefiting you?
We are able to easily ingest, transform, display, and train on all our data, in one place.
Databricks for all
What do you like best about the product?
Is a plataform for everyone, no matter what role you have, Databricks has build something for you!
What do you dislike about the product?
In terms of governance to add more federation sources to be a máster catalog not only for Databricks assets but for application layer or reporting!
What problems is the product solving and how is that benefiting you?
Time to deployment, ETL and orchestration strategy.
Great platform to try, train and infer your AI models!
What do you like best about the product?
Ease to use platform
Provides foundational and popular AI models
Allows experimenting, training and inferring
Provides foundational and popular AI models
Allows experimenting, training and inferring
What do you dislike about the product?
Documentation could be better. It's not intuitive to find the data intelligence platform section.
What problems is the product solving and how is that benefiting you?
It allows for training, fine-tuning and inferring AI models which is helping us a lot to try the AI models.
Databricks is awesome
What do you like best about the product?
Combining data, analytics, AI and BI into a single platform
What do you dislike about the product?
Quicker speed from preview to General availability
What problems is the product solving and how is that benefiting you?
Proving best in class audit results
Best tool ever
What do you like best about the product?
Flexible and user friendly, business user empowered
What do you dislike about the product?
Reluctance from IT to fully empower business users
What problems is the product solving and how is that benefiting you?
Unification of financial metrics
Data, all in one place
What do you like best about the product?
Ability to use all data in a single
Platform
Platform
What do you dislike about the product?
Too many features remain in preview, need faster turn around
What problems is the product solving and how is that benefiting you?
Benchmarking
showing 1 - 10