Sr. Data Scientist at a computer software company with 1,001-5,000 employees
Real User
Top 10
Jun 27, 2026
I cannot really say much about needed improvements for TigerGraph in my current work. I think it has great adaptation with AI. If I had to pick at something, it might be more AI integration, but overall, it is hard to suggest improvements as TigerGraph supports many technologies well.
I feel that more enablement could be improved for better understanding, which is an area where I see room for enhancement. It primarily needs more instructor-led training associated with TigerGraph University to enable broader usage and certifications, allowing more users to utilize its features effectively. I chose a rating of eight out of ten mainly because documentation, more demos, and improved customer approachability still present opportunities for improvement. Having more projects using TigerGraph or additional use cases could potentially justify a perfect ten. TigerGraph is well-equipped so far, but there could be enhancements in cost-wise partner enablement initiatives, aiming to reduce costs for implementations through AWS, Azure, or Google Cloud platforms.
Every organization needs improvements. GraphStudio has UI/UX issues and bugs. The visual interface has several frustrating bugs, for example, graph exploration results sometimes disappear and node selection can be difficult when edges are very close together. Additionally, loading jobs created by GSQL do not appear in GraphStudio at all, forcing users to manage the UI and command line separately. A dark theme is also missing in the new versions. Regarding query development, the query installation process can be painfully slow, with some users reporting that installed queries get dropped or hang, especially when the system is overloaded. Additionally, there is no way to create or replace a query directly, which complicates iterative development. Large result sets over two gigabytes cannot be paginated, requiring workarounds such as writing to files. There are some missing modern developer features such as query profiling, and TigerGraph does not support openCypher syntax or interpret queries, and it failed to track create, update, and delete operations. The lack of built-in pagination for large REST API responses remains a pain point for building production applications. Integrating TigerGraph with popular Graph RAG frameworks has been challenging. The light-RAG integration suffers from performance issues due to inefficient per-node operations rather than batch processing. Microsoft's GraphRAG does not abstract the storage layer, so its output can be stored independently.
Senior DB Engineer And Sre at a tech vendor with 10,001+ employees
Real User
Top 20
May 14, 2026
TigerGraph can be improved by adding features for multi-updates and in-place upgrades when documents are inserted. Additionally, it should enhance scaling capabilities as data grows, with more collections and documents added. The performance of complex joins should improve to make relationships more direct instead of requiring multiple hops. Beyond those improvements, I suggest increasing visibility on internal features, more metrics, and views to help identify potential issues.
Responsable Del Equipo D Más D at a tech services company with 51-200 employees
Real User
Top 10
May 12, 2026
TigerGraph could improve by making the community version more complete and not limited to a monolithic service. Additionally, integrating models well would be beneficial.
TigerGraph can improve on certain factors, particularly the simple query language, as the learning curve can be very hard for new users or beginners. The visualization tools could also be improved. For new developers, especially those who are freshers, the learning curve for the simple query language should be made easier because it is relatively harder for them to learn without much experience in any tech stack. I have a few team members who are freshers, and it is relatively harder for them to learn this kind of solution.
TigerGraph offers a graph analytics platform that efficiently handles large-scale and complex data relationships, providing insights for informed decision-making.Specialized for big data, TigerGraph leverages a native parallel graph architecture to analyze data relationships rapidly. It is designed to manage extensive datasets, providing real-time insights that are invaluable for sectors like financial services, healthcare, and telecommunications. With its scalable infrastructure, it supports...
I cannot really say much about needed improvements for TigerGraph in my current work. I think it has great adaptation with AI. If I had to pick at something, it might be more AI integration, but overall, it is hard to suggest improvements as TigerGraph supports many technologies well.
I feel that more enablement could be improved for better understanding, which is an area where I see room for enhancement. It primarily needs more instructor-led training associated with TigerGraph University to enable broader usage and certifications, allowing more users to utilize its features effectively. I chose a rating of eight out of ten mainly because documentation, more demos, and improved customer approachability still present opportunities for improvement. Having more projects using TigerGraph or additional use cases could potentially justify a perfect ten. TigerGraph is well-equipped so far, but there could be enhancements in cost-wise partner enablement initiatives, aiming to reduce costs for implementations through AWS, Azure, or Google Cloud platforms.
Every organization needs improvements. GraphStudio has UI/UX issues and bugs. The visual interface has several frustrating bugs, for example, graph exploration results sometimes disappear and node selection can be difficult when edges are very close together. Additionally, loading jobs created by GSQL do not appear in GraphStudio at all, forcing users to manage the UI and command line separately. A dark theme is also missing in the new versions. Regarding query development, the query installation process can be painfully slow, with some users reporting that installed queries get dropped or hang, especially when the system is overloaded. Additionally, there is no way to create or replace a query directly, which complicates iterative development. Large result sets over two gigabytes cannot be paginated, requiring workarounds such as writing to files. There are some missing modern developer features such as query profiling, and TigerGraph does not support openCypher syntax or interpret queries, and it failed to track create, update, and delete operations. The lack of built-in pagination for large REST API responses remains a pain point for building production applications. Integrating TigerGraph with popular Graph RAG frameworks has been challenging. The light-RAG integration suffers from performance issues due to inefficient per-node operations rather than batch processing. Microsoft's GraphRAG does not abstract the storage layer, so its output can be stored independently.
TigerGraph can be improved by adding features for multi-updates and in-place upgrades when documents are inserted. Additionally, it should enhance scaling capabilities as data grows, with more collections and documents added. The performance of complex joins should improve to make relationships more direct instead of requiring multiple hops. Beyond those improvements, I suggest increasing visibility on internal features, more metrics, and views to help identify potential issues.
TigerGraph could improve by making the community version more complete and not limited to a monolithic service. Additionally, integrating models well would be beneficial.
TigerGraph can improve on certain factors, particularly the simple query language, as the learning curve can be very hard for new users or beginners. The visualization tools could also be improved. For new developers, especially those who are freshers, the learning curve for the simple query language should be made easier because it is relatively harder for them to learn without much experience in any tech stack. I have a few team members who are freshers, and it is relatively harder for them to learn this kind of solution.