Unified data engineering has streamlined supplier scorecards and operational analytics
What is our primary use case?
My main use case for Palantir Foundry is from the data engineering perspective.
A specific example of how I use Palantir Foundry for data engineering involves raw data stored in Redshift AWS, which we are using those tables in the form of a dataset in Foundry. We are ingesting that data into Foundry and using it for cleaning purposes. After cleaning the data, we create Ontology objects and use them for operational applications in the Workshop.
One of the use cases that I found with Palantir Foundry is when I worked on the supplier scorecard, which is dedicated to understanding supplier reviews based on the goods supplied. The company assigns ratings to their products through a supplier scorecard, providing scores to their suppliers. We used multiple datasets and created objects, adding our own logic in the Code Repository to check supplier goods by percentage and count, generating aggregated values in the Workshop app. Based on these parameters, business management can make decisions and take actions to update the supplier's score.
What is most valuable?
Palantir Foundry offers great features, including data connection specifically for ingesting data from various third-party servers like AWS or Azure, with around 250 plus data connections available. Additionally, it includes Pipeline Builder, one of the best ETL tools for transforming data from raw to gold layer data, following a medallion architecture of bronze, silver, and gold. In more complex use cases, the Code Repository offers a fully code-based solution while Pipeline Builder serves as a no-code, low-code tool; my preference leans towards Pipeline Builder for data refinement.
For data analytics, it features Contour, allowing for data analysis, and ontology objects for creating links between multiple objects with actions for CRUD operations throughout the Workshop. It also has Quiver for exploring objects using AI tools, enabling business users to ask questions in their native language, which Quiver converts into queries for report generation. Another significant feature is AIP Logic, akin to agentic AI, processing existing data with multiple AI models. AIFTA is another cool feature that needs only a prompt to handle various tasks, such as creating ETL pipelines based on raw data stored, selecting to create the pipeline in either Pipeline Builder or Code Repository as needed, and also supporting object creation at the branch level.
In the Pipeline Builder, we can use Databricks as a compute profile, which is one of the coolest features. The OSDK is a new feature that allows creating custom UI pages using React or Angular, fetching data through API, should the business be unsatisfied with existing widgets in the Workshop. There is also MCP Hub, which uses Model Context Protocol to operate Palantir Foundry from a local machine using LLMs, generating and deploying code efficiently.
Palantir Foundry has positively impacted my organization through multiple use cases, such as warranty data refinement, where previously we struggled with identifying the number of claims related to specific products. Analyzing the data helped us craft a Workshop application that tracks claims by country and product, enabling report generation for management action on defective products.
What needs improvement?
I believe Palantir Foundry could improve by introducing a tool to restrict object-level creation to specific people, such as developers. A dedicated application could streamline requests for access to data across different organizational verticals, enabling better tracking of costs associated with specific use cases and improving identification of data access requests.
Regarding documentation, I find that when I face issues, the outdated documentation is not helpful; for example, while trying to create a webhook to fetch SharePoint metadata, I found available resources lacking relevance, needing significant updates to assist users properly.
For how long have I used the solution?
I have been using Palantir Foundry for the last four years.
What do I think about the stability of the solution?
Palantir Foundry is stable.
What do I think about the scalability of the solution?
Regarding scalability, if you have billions and trillions of records, Palantir Foundry accommodates ETL pipelines with a dedicated compute profile that lets you choose from several sizes, whether small, large, extra-large, or custom to fit your needs.
How are customer service and support?
Customer support is really good; when we encounter issues, raising a ticket with a screenshot leads to responses typically within a week or twice a month, depending on their organization.
Which solution did I use previously and why did I switch?
Previously, for data engineering, we used Databricks; however, it lacked the capabilities we found in Palantir Foundry, which allow for analysis, reporting, and automation without needing to implement additional functions such as AWS Lambda. I appreciate that Palantir Foundry offers dedicated automation tools significantly simplifying processes.
What was our ROI?
We have seen a return on investment, primarily saving money on developers. With traditional development requiring many specialized roles, Palantir Foundry allows us to operate efficiently with fewer personnel, often enabling just a couple of front-end developers to manage our processes, thus noticeably reducing time and costs when completed effectively.
What's my experience with pricing, setup cost, and licensing?
My experience with pricing, setup cost, and licensing has not been too overwhelming; I worked closely with a management colleague who explained how they check for cost based on user activity and individual vertical usage.
Which other solutions did I evaluate?
I primarily evaluated Databricks, which I found lacking compared to Palantir Foundry's robust offerings.
What other advice do I have?
My advice for others considering Palantir Foundry is that it delivers an ecosystem eliminating the need for third-party applications, greatly simplifying tasks without requiring extensive efforts in model training or other processes, making it a strong option for organizations. Security and data governance are also significant advantages. I have covered everything I know and have used regarding Palantir Foundry. I would rate this product a ten out of ten.
Which deployment model are you using for this solution?
Private Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Amazon Web Services (AWS)
Palantir Foundry: Seamlessly Integrating AI Workflows into Our Data Ecosystem
What do you like best about the product?
Palantir Foundry has allowed us to incorporate AI workflows into our tech stack and data ecosystem.
What do you dislike about the product?
It's not cheap - but at least you get what you pay for!
What problems is the product solving and how is that benefiting you?
It helps us manage our data ecosystem while also connecting it to AI, helping us identify insights in our data that we can turn into actions for our business.
Good place for softwares
What do you like best about the product?
Good experience,Ease of Use,easily implementing products,good support,integrating with other products. And been using the products frequently
What do you dislike about the product?
Nothing as such until now, been using since many days
What problems is the product solving and how is that benefiting you?
Helping me to make the automation with machine learning models to make the predictive analysis very quick almost accurate.
A great tool for viz
What do you like best about the product?
It provides a great eco system to manage data and analyze it
What do you dislike about the product?
Nothing as such so far. SInce I am new to this.
What problems is the product solving and how is that benefiting you?
It itegrates all problems and gives a better solution
Best Review
What do you like best about the product?
I loved about how it's delivered it's most functional ways to create more obsessities its can be used to used on many things that related to many types of tasks.
What do you dislike about the product?
It's something of the functions are very pitty never had something like this before which has very different types of disadvantages unlike the other softwares.
What problems is the product solving and how is that benefiting you?
It's benefiting me with my business it's can be used to solved many financial and functional things that can be in many ways has solved the problems on my business.
State of the art data management
What do you like best about the product?
The interface looks great. A lot of information can be viewed, linked and stored. The data set protection (only post treating functions are editable) is a real plus !
What do you dislike about the product?
Maybe too « informatics » oriented yet. Applications can be very slow. Chart bars with 2 colors stripes cannot be displayed (which is a basic function of PowerPoint !)
What problems is the product solving and how is that benefiting you?
Big data analysis, producing indicators automatically and using applications to get datas (+searching)
Recommendations to others considering the product:
Make it available to all user (as excel), simple to use and develop applications.
