Reviews from AWS Marketplace
0 AWS reviews
-
5 star0
-
4 star0
-
3 star0
-
2 star0
-
1 star0
External reviews
External reviews are not included in the AWS star rating for the product.
Super powerful and simple
What do you like best about the product?
WhyLabs is so useful for debugging, maintaining, and developing production ML systems. It's portable, highly configurable, and provides unified visibility across a broad range of systems/services in a production ML ecosystem. It dramatically improves ML devops and is incredibly easy to get started. It integrates very easily with all the systems I've used, implementation is a breeze, and there's very low friction to get started. I've caught tons of issues using it.
What do you dislike about the product?
I don't really have any complaints about it, it does its thing incredibly well for my use cases.
What problems is the product solving and how is that benefiting you?
Visibility and observability that is unified across all stages of ML dev and ops: model training, development, deployment, testing, etc. Importantly it integrates with a broad range of systems/languages.
It accelerates development/deployment by provide great visibility.
For ops, it has caught so many super costly issues which everything else misses, especially things related to train/serve skew or semantic drift.
Seriously, it's a must-have for productionizing ML.
It accelerates development/deployment by provide great visibility.
For ops, it has caught so many super costly issues which everything else misses, especially things related to train/serve skew or semantic drift.
Seriously, it's a must-have for productionizing ML.
- Leave a Comment |
- Mark review as helpful
WhyLabs answers Why
What do you like best about the product?
WhyLabs Al is easy to understand and use. It enhances interpretability, making it easier for data scientists and machine learning engineers to understand model behavior
What do you dislike about the product?
More detailed dashboard would be more helpful
What problems is the product solving and how is that benefiting you?
It enable me monitoring and understanding the interaction between my data and model
Intuitive and user-friendly product.
What do you like best about the product?
I see so much potential how it can be applied to my own academic work. It's very friendly by providing easy-to-read plots and intuitive platform.
What do you dislike about the product?
I found some of the features very slow to display.
What problems is the product solving and how is that benefiting you?
I don't have any specific problem at the moment, but I see potential in using it to conduct a short text analysis as an exploratory method.
Amazing 1:1 Technical Support
What do you like best about the product?
While we have not gone extremely deep into all the features that Whylabs provides, we have had extraordinary 1:1 technical support. They are quick to respond, fast at finding solutions, and eager to make improvements for your use case.
What do you dislike about the product?
There is some complexity setting up data that lags behind the current date but with some changes you can easily overcome this.
What problems is the product solving and how is that benefiting you?
Whylabs helps us monitor our models in production.
A robust AI Observability platform with a thoughtful approach to data privacy
What do you like best about the product?
As an AI consulting and services firm, we’re always on the lookout for tools that our customers can leverage to improve their adoption of AI. Model monitoring and maintenance are big needs for any company actively investing in AI that wants to turn those investments into returns.
We’ve been testing out WhyLabs as a potential observability solution and have been impressed with its capabilities. Their recent platform expansion with LLM observability and guardrails aligns well with the Gen AI capabilities that are becoming important to many of our customers.
WhyLabs offers a wide range of functionality, including data profiling, model monitoring, alerting, and root cause analysis. The platform covers many of the end-to-end workflow needs of an ML team.
A key part of their approach is the open-source whylogs library for local data profiling. Rather than sending raw data to the cloud, whylogs lets you summarize on-premises before sending compact statistical profiles. Contrary to some other monitoring solutions, this aligns very well with many of our clients’ strict data privacy expectations.
whylogs itself is straightforward to integrate and use. You can add monitoring to a pipeline with minimal fuss. As an open-source tool, it provides transparency into what’s being tracked and how.
On the support side, we’ve found the WhyLabs team extremely nice and responsive. The platform documentation is fairly comprehensive as well. Another key plus is that we found their pricing model really fair.
For teams looking for an observability solution, WhyLabs is definitely worth a close look. The platform covers a wide range of ML workflow needs, and their approach to data privacy is a plus. We look forward to using it more broadly!
We’ve been testing out WhyLabs as a potential observability solution and have been impressed with its capabilities. Their recent platform expansion with LLM observability and guardrails aligns well with the Gen AI capabilities that are becoming important to many of our customers.
WhyLabs offers a wide range of functionality, including data profiling, model monitoring, alerting, and root cause analysis. The platform covers many of the end-to-end workflow needs of an ML team.
A key part of their approach is the open-source whylogs library for local data profiling. Rather than sending raw data to the cloud, whylogs lets you summarize on-premises before sending compact statistical profiles. Contrary to some other monitoring solutions, this aligns very well with many of our clients’ strict data privacy expectations.
whylogs itself is straightforward to integrate and use. You can add monitoring to a pipeline with minimal fuss. As an open-source tool, it provides transparency into what’s being tracked and how.
On the support side, we’ve found the WhyLabs team extremely nice and responsive. The platform documentation is fairly comprehensive as well. Another key plus is that we found their pricing model really fair.
For teams looking for an observability solution, WhyLabs is definitely worth a close look. The platform covers a wide range of ML workflow needs, and their approach to data privacy is a plus. We look forward to using it more broadly!
What do you dislike about the product?
Nothing major. There is always room for improving documentation and platform features.
What problems is the product solving and how is that benefiting you?
As an AI consulting and services firm, Tryolabs' mission is to help companies accelerate their adoption of AI. We guide clients through their AI roadmaps, building, deploying, scaling and monitoring production-ready ML systems.
Robust monitoring and observability are crucial to running AI responsibly in production. WhyLabs provides an invaluable end-to-end solution for the monitoring challenges ML teams face.
WhyLabs gives us comprehensive observability into our clients' ML pipelines. The privacy-preserving design using whylogs is also critical for many of our clients.
By leveraging WhyLabs, we can focus on rapidly delivering AI value for our clients rather than building custom monitoring. It's a trusted platform that sets up our clients for ongoing success after models go into production.
Robust monitoring and observability are crucial to running AI responsibly in production. WhyLabs provides an invaluable end-to-end solution for the monitoring challenges ML teams face.
WhyLabs gives us comprehensive observability into our clients' ML pipelines. The privacy-preserving design using whylogs is also critical for many of our clients.
By leveraging WhyLabs, we can focus on rapidly delivering AI value for our clients rather than building custom monitoring. It's a trusted platform that sets up our clients for ongoing success after models go into production.
A must have if you're organization is mature enough to use it
What do you like best about the product?
The core functionality is explicitly important when you are having AI/ML solutions running on production. Instead of spending time on manual checks (that can be prone to error) you can set up monitors to inform you if something is going wrong. It is flexible enough to build your cusotm metric and monitors which allows for versatile usage.
What is highly important to me is the contact with WhyLabs team. Their support is superb which results in fact that even if something is difficult to implement, I know I have people who will help me get the best out of the platform.
What is highly important to me is the contact with WhyLabs team. Their support is superb which results in fact that even if something is difficult to implement, I know I have people who will help me get the best out of the platform.
What do you dislike about the product?
Building custom monitors and metrics is a chellanging experience. I struggle sometimes with code issues so better FAQ or documentation could be helpfull to avoid them.
What problems is the product solving and how is that benefiting you?
We are having multiple user facing projects that are working on production. We need to know if their performance is deteriorating, if we see some bias, data drift or missing data.
Great product on the forefront of AI/ML observability, including LLMs
What do you like best about the product?
WhyLabs is a great team building an innovative product at the forefront of AI/ML. I love the OSS approach and have been using and recomending their new product Langkit to my LLM dev community. We all have a lot of questions about how we're going to measure and protect our LLM-based products. Very cool that Whylabs has come oout with a comepelling product so quickly, while the ground is still shifting.
What do you dislike about the product?
It is still early days and there is a lot more to build, measure and monitor, especially as the LLM space contnues to rapidly evolve. I am looking forward to seeing new features and capabilities in Langkit as the models get more sophisticated and our underdstanding of the technical and business metrics emerge that will govern how we build LLM-based products and services.
What problems is the product solving and how is that benefiting you?
My colleagues and I need to understand and keep our finger on the pulse of these currently still unweildy beasts (LLMs) that don't always act in a predictable manner. Output quality, relevance and security are still in its infancy wrt LLMs. WhyLab's Langkit is off to a great start giving us visibiity into these crucial metrics, giving us the confidence to move forward with our own LLM-based products and services.
WhyLabs made monitoring our generative models flexible and useful
What do you like best about the product?
The platform is flexible, privacy friendly (you don't have to send user data) and provides a great command center view with many alerting options.
What do you dislike about the product?
The software is great, no complaints at all.
What problems is the product solving and how is that benefiting you?
WhyLabs helps us monitor our models to ensure the performance doesn't change over time, that any upstream or downstream data inconsistencies or changes are noticed quickly. We monitor some simple metrics, for example word count, and some more complex metrics, such as cosine similarity. I like that the tool is flexible enough to monitor anything we can calculate, then provides best practices for alerting around unexpected variance and extreme values.
Excellent product by an amazing team
What do you like best about the product?
WhyLabs is a great solution for our team in monitoring ML features. With WhyLabs seamless integrated in our daily pipelines, we can keep track of our data quality, receive timely alerts via both email and Slack, and detect the anomalies in their early stages.
We are also impressed by the tremendous support from the WhyLabs team. Collaborating with them has been an absolute delight.
We are also impressed by the tremendous support from the WhyLabs team. Collaborating with them has been an absolute delight.
What do you dislike about the product?
Just like with any new product, there is always room for improvement in both the platform and its monitoring capabilities.
I am anticipating new features in dashboards that offer data profiles at more granular levels and from different perspectives. Additionally, having a comprehensive API client tool to check and retrieve profiles would benefit our specific use cases.
I am anticipating new features in dashboards that offer data profiles at more granular levels and from different perspectives. Additionally, having a comprehensive API client tool to check and retrieve profiles would benefit our specific use cases.
What problems is the product solving and how is that benefiting you?
Delivering consistent and reliable data to our customers stands as a fundamental responsibility of our team. However, given the chaos in the real world and the complexity of infrastructure dependencies, detecting data quality issues promptly poses notable challenges. WhyLabs provides solutions to monitor statistical properties of large datasets, giving us the confidence to move forward with our data services.
Outstanding Support and Speed of Innovation
What do you like best about the product?
We recently integrated the WhyLabs AI Observatory into our team's model serving platform to monitor input features and model scores in near real-time. I was impressed by its seamless integration and range of capabilities. They provided some good examples, so it was easy to integrate monitoring of our data.
I also like the large amount of monitors and dashboards to debug various aspect of ML data and models. We set up alerts through Slack which was also easy to set up. The alerts are useful to detect issues with our model's features such missing data. We also watch WhyLabs dashboards when deploying new model versions to ensure a smooth rollout.
Another highlights is is user interface. It's user-friendly, clear, and intuitive. They have some preset monitors that are easy to add with just a few clicks. I noticed they added more preset monitors recently which shows their rapid innovation.
Finally, and what I like the most, is its customer support. The team is responsive, and they are always ready to assit, and they quickly reply for any questions or help we ask.
I also like the large amount of monitors and dashboards to debug various aspect of ML data and models. We set up alerts through Slack which was also easy to set up. The alerts are useful to detect issues with our model's features such missing data. We also watch WhyLabs dashboards when deploying new model versions to ensure a smooth rollout.
Another highlights is is user interface. It's user-friendly, clear, and intuitive. They have some preset monitors that are easy to add with just a few clicks. I noticed they added more preset monitors recently which shows their rapid innovation.
Finally, and what I like the most, is its customer support. The team is responsive, and they are always ready to assit, and they quickly reply for any questions or help we ask.
What do you dislike about the product?
One point to mention is the discoverability and enablement of some of their advanced features. However, this could be a results of their rapid innovation, which continuously expands the platform's capabilities. A bit more documentation on navigating these features could improve the user experience, but it's a small aspect in an otherwise impressive system.
What problems is the product solving and how is that benefiting you?
WhyLabs AI Observatory is a crucial tool that helps us to maintain the integrity and performance of our real-time Machine Learning models. It lets us monitor the features and model predictions to ensure they behave as per our expectations.
By integrating WhyLabs, we are able to comprehensively capture potential data drift in predictions and detect anomalies in model inputs and ouputs, which significantly increases the reliability of our models. We have also established monotors that promptly alerts us to any emerging issues, enabling swift and appropriate action.
By integrating WhyLabs, we are able to comprehensively capture potential data drift in predictions and detect anomalies in model inputs and ouputs, which significantly increases the reliability of our models. We have also established monotors that promptly alerts us to any emerging issues, enabling swift and appropriate action.
showing 11 - 20