Teleport has been a solid and secure access management solution, but there are a few areas that could be improved, particularly around the initial setup and RBAC configuration. Beyond onboarding and RBAC complexity, there are areas that could be enhanced from an integration perspective. While SSO integration works well, deeper out-of-the-box integrations with SIEM and monitoring tools would be beneficial.
I think since nowadays it is more about AI, Teleport might need to improve or focus more on how to interact with MCP agents, provisioning access to non-human agents to connect to back-end resources such as databases or any data sources, for example, Snowflake. There should be some granularity on it because many threats right now, even advanced threats, are using AI. I haven't tried that part yet, but I look forward to Teleport improving access to the AI environment.
Regarding how Teleport can be improved, I believe we can improve by using more automation on all devices. Sometimes our devices do not reflect the error, and then we work on why it is not showing any alarm or issue, even though we experience distortion or issues on the TS. We work to resolve these issues by changing some devices and their alarms. Our development team consults with our R&D team regarding these types of issues to see why we are not getting device alerts. Then we upgrade the devices, firmwares, and fix bugs.
Test lead at a outsourcing company with 5,001-10,000 employees
Real User
Top 10
Jan 19, 2026
Managing 700 to 800 services is not easy. Every day, we are facing many challenges. For example, sometimes we face multiple incidents on multiple channels. We are also working on how to improve major faults.
While the setup for installation with Teleport Helm chart or on a Raspberry Pi is smooth, I believe they need to streamline the installation process further, especially on Kubernetes clusters or regular computers. Based on my experiences testing the product, I think more automation regarding configuration verification is needed as some specifications documented do not align with practical requirements. The learning curve is an area for improvement.
I cannot download or share session recordings directly with others, which would be useful for documentation or reviews. Additionally, while the backend worker can be upgraded, there is a limitation with scaling, as it supports only a single node even in Kubernetes.
Teleport streamlines access management for engineering teams, enhancing security and operational efficiency by implementing robust role-based access controls and audit capabilities.
Teleport is designed to address the complexities of modern IT environments by providing a comprehensive platform that unifies access controls for SSH, Kubernetes, databases, and web applications. With a focus on security and compliance, it offers end-to-end encryption and ensures secure access points while...
Teleport has been a solid and secure access management solution, but there are a few areas that could be improved, particularly around the initial setup and RBAC configuration. Beyond onboarding and RBAC complexity, there are areas that could be enhanced from an integration perspective. While SSO integration works well, deeper out-of-the-box integrations with SIEM and monitoring tools would be beneficial.
I think since nowadays it is more about AI, Teleport might need to improve or focus more on how to interact with MCP agents, provisioning access to non-human agents to connect to back-end resources such as databases or any data sources, for example, Snowflake. There should be some granularity on it because many threats right now, even advanced threats, are using AI. I haven't tried that part yet, but I look forward to Teleport improving access to the AI environment.
Regarding how Teleport can be improved, I believe we can improve by using more automation on all devices. Sometimes our devices do not reflect the error, and then we work on why it is not showing any alarm or issue, even though we experience distortion or issues on the TS. We work to resolve these issues by changing some devices and their alarms. Our development team consults with our R&D team regarding these types of issues to see why we are not getting device alerts. Then we upgrade the devices, firmwares, and fix bugs.
Managing 700 to 800 services is not easy. Every day, we are facing many challenges. For example, sometimes we face multiple incidents on multiple channels. We are also working on how to improve major faults.
While the setup for installation with Teleport Helm chart or on a Raspberry Pi is smooth, I believe they need to streamline the installation process further, especially on Kubernetes clusters or regular computers. Based on my experiences testing the product, I think more automation regarding configuration verification is needed as some specifications documented do not align with practical requirements. The learning curve is an area for improvement.
I cannot download or share session recordings directly with others, which would be useful for documentation or reviews. Additionally, while the backend worker can be upgraded, there is a limitation with scaling, as it supports only a single node even in Kubernetes.