Whether in-house or outsourced, we have the right plan for you.
Whether in-house or outsourced, we have the right plan for you.
Ango Hub is the data labeling tool for annotation projects of any size.
On the cloud or on your own premises.
Sign up on our website and start labeling right now. Limited to 5 users and 10k annotations.
Free, no credit card required.
Just log in and go. Invite your whole team, create projects, annotate, and manage your workflow.
Set up Ango Hub on your own server. Nothing leaves your datacenter, and you control the process end-to-end.
Ango Hub is the versatile data annotation platform for projects of any size.
Yes, Ango Hub is built to allow you and your team to perform annotation tasks of any type and size. In the Cloud and On-Prem plans, there are no limits to how many people and assets you can have in your labeling projects.
Yes, we provide a fully-managed data labeling service with an immediately scalable workforce, for projects of any size and difficulty. Check out our Ango Service page for more information.
Yes, our Ango Hub On-Prem package is what you’re looking for. We set up Ango Hub on your server in less than 15 minutes (we use Kubernetes) and you can start labeling immediately with your own team. If you need to scale, upon request, our fully-managed labeling workforce can connect to your server via VPN and help you to label your data.
We support text, audio, image, video, multi-page images, and native PDF annotation.
Check out our Ango Hub page for a complete list of our labeling tools, but shortly, everything your team may need, from the usual points and polygons to more esoteric rotated bounding boxes, tables, and more.
Yes. We have a free plan with which you can try out the platform for yourself. Simply go to hub.angoai.basework.studio/dev and sign up now, no credit card required. Limited to five users, 2000 lifetime assets, and 10k lifetime annotations.
No. Users of our Ango Hub On-Prem plan can tap into Ango Hub’s ever growing fully-managed labeling workforce. If you ever need to scale your team quickly, or if you want to switch to outsourcing your data labeling tasks, contact us and our annotators will connect to your platform via VPN, and become part of your labeling team.
Yes. Built into Ango Hub are consensus and review features helping to make sure the quality of your labels stays high. You can have up to 30 people annotating the same asset: Ango Hub will calculate the consensus between them and tell you whether or not the labels agree. Plus, you can have a reviewer double checking the result.
Users of our Free and Cloud plans have two options: they can import the data to Ango Hub (therefore to our cloud servers) or they can directly annotate data they store on a cloud storage service (e.g. S3, R2, GCP.) In this second case, your data does not get uploaded to our servers and it remains stored in your cloud storage of choice.
Users of our On-Prem package can choose to label data directly from their server, or to import data they are storing on a cloud storage service. In both cases we have no access to the data.
Let Ango AI do all the heavy lifting.
We get your data, annotate it, and send you the labels. It’s that simple.
Everything you need to have your data labeled. Our remote workforce will connect to our servers and annotate your data for you.
Our annotators label your data from our secure facility with 24/7 CCTV with no devices allowed, logged entries, and more.
The software is installed on your datacenter, and label with your own team. Then, tap into our workforce through VPN when you need.
Ango Service is the quality-centric data annotation service provided by Ango AI. You send us your data, tell us how to label it, and get your annotations. It’s as simple as that.
As soon as you send us your data and instructions on how to label it, a dedicated project manager will educate our fully-managed labeling workforce on how to annotate it, then distribute the data to them. Labelers then use our own platform Ango Hub to perform the annotations.
If requested by the customer, we can have the same asset labeled by more than one person, and/or add a review step to ensure maximum annotation quality.
Throughout the whole process, you can download the labels that have been completed so far and give us feedback. We also run automated QA processes to make sure your annotations are of the maximum quality.
Yes. We pride ourselves in our full transparency. Throughout the entire process, our customers can see how many labels have been completed, when, by who, how long each person took to label each asset, and much more.
We charge a fixed amount by the asset, then a fixed amount by the label. We do not price based on the amount of time spent by our labelers on the project. This ensures that you know the price of the entire project beforehand without surprises. Contact us for more information on our prices.
Yes. They use Ango Hub, the same software we also provide to our customers who want to do labeling in-house. This means you can check out Ango Hub’s features on our Ango Hub page, and sign up for free on hub.angoai.basework.studio/dev to see it yourself.
There are no limits to how you can send us your data. We have worked with methods ranging from FTP and cloud storage to encrypted hard drives shipped to our address. Whatever the way you wish to send us your data, we support it.
Your data is your data, and security is at the foundation of everything that we do. Before seeing any of your data, every person who will see it in any way signs a comprehensive NDA. Our annotators, project managers, and developers are trained in security best practices from day one.
For customers of our Essential plan, we store your data on our encrypted cloud, and annotators connect to it via an encrypted connection. For customers of our Secure plan, your data is stored locally in our datacenter, which is not connected to the internet. Labeling is done internally within our intranet, in a room where no personal electronic devices are allowed, monitored 24/7 by CCTV, and with access limited to authorized personnel.
Firstly, we assign a project manager to your labeling tasks, whose responsibility is to ensure that your labels are of the highest quality. As our annotators label, we run the labels through automated and semi-automated QA processes to detect outliers and bad labels. Bad labels are then deleted, and the related asset is put back into the labeling pool.
We employ full transparency during the labeling process, ensuring that you are able to see the labeling progress at any time as well as a number of metrics regarding the quality of the work performed by our labelers, such as their consensus score, time spent on each individual label, and more.
We then provide a consensus system, where more than one person labels the same data point, as well as an additional review process. Our platform was built from the ground up to ensure label quality, providing features such as real-time troubleshooting, centralized labeling instructions, and more.