Isahit is the first ethical data labeling solution for AI and data processing in Europe. Thanks to a trained, supervised and diversified workforce, an agile team as well as a very demanding quality and security review, isahit deals with more than 300 customers, in France and internationally, including many major accounts from CAC 40 (Airbus, Sodexo, L'Oréal, E.Leclerc, IBM...).
Driven by the desire to promote social inclusion through work and training, isahit mission is to offer to more than 10,000 people around the world the opportunity to work, on a daily basis, and train in the digital sector through its online platform by 2025.
In order to ensure that each project runs smoothly, isahit sets up specific and precise processes. A successful annotation project implies an iteration and a validation phase between the client and isahit.
These iteration phases, also called test phases, will allow the contributors who work on our customers' projects to practice on test data until they have a complete understanding of the project and a knowledge of all the scenarios (edge cases). These iteration phases are reinforced by dedicated and personalized training and precise quality control processes to guarantee high quality data processing and total scalability.
The success of our projects and the satisfaction of our clients is based on an extremely rigorous quality review that we reinforce year after year.
The quality assurance of the various projects is based on different points:
- The training of our contributors on each project and different technologies. In order to ensure complete and well-executed training, we have developed the "Isahit Digital Academy" which offers specific training modules for each project. More than 15,000 hours of training have been provided since isahit was founded in 2017.
- Clear and detailed instructions. The project manager provides written and video instructions, and presents a large number of samples and examples to the contributors. In addition, each project manager remains available to support contributors and answer questions.
- To ensure that contributors understand the needs of the various projects, tests and FAQs are set up. Throughout the project, our teams perform controls on the various tasks of the contributors in order to ensure quality control on the entire project
- Daily feedback and exchange sessions are set up to make sure the information flows smoothly.
- Performance indicators are defined and monitored to evaluate in real time the productivity of contributors and the progress of the team
The purpose of isahit is to encourage social inclusion through employment and training. Today, we are proud to have been able to support nearly 2000 women by offering them a new form of remote and flexible work, training and a community to rely on and grow.
Our community of contributors, called HITers (HIT=human intelligence task), come from economically fragile countries and are looking for an additional income to finance their studies or an entrepreneurial project. We are very happy to be able to rely on a very rich and diverse community of women who come from 39 different countries with different languages and cultures.Thanks to this rich cultural diversity, we are able to meet the different needs of our clients, handle projects worldwide and prevent the different biases inherent to artificial intelligence.
For natural language processing projects, for example, we can work in over 20 languages.
Isahit recruitment process is transparent. The evaluation of our contributors is based on their language skills, their business expertise, their localization and their labeling annotation background on certain projects. We only work with people who hold a self-employed status in their country and are legally entitled to work on our platform.
Project and contributor management is affiliated with a dedicated project manager who is responsible for understanding the client's needs and guidelines and then properly training our contributors accordingly. He or she sets milestones and ensures that the team has the required resources to achieve them. If necessary, he/she will propose new training to the workforce using different processes available on our platform. Our project managers are based in Paris and can be reached throughout the day.
Depending on the project, we will set up a team of contributors with the right skills. Thanks to our assignment algorithm, we select the right contributors without any form of favoritism but solely on the basis of their professional, academic, linguistic or cultural backgrounds.
Security and privacy are a major issue in AI data processing. We have strengthened this parameter to the maximum to guarantee our customers total protection of their data. From data security processes for our contributors, to the remote working policy in place and the RGPD compliance of our contracts - but also to the secure storage of data in the cloud thanks to TLS/SSL encryption and vulnerability testing, isahit ensures optimal security for its customers.
Now that you know us a little better as a data labeling partner, make an appointment with our team who will be happy to discuss your various needs with you!
Join our 300 customers and our 1200 active contributors! Together let's make Tech and AI a real job and training opportunity for women around the world!
Winners of the David with Goliath competition alongside our partner L'Oréal, we had the opportunity to broadcast an advertising campaign in the Parisian subways and train stations! Discover the story behind our ad...
Choosing the right data annotation partner can be a long and tedious job, which is why we have outlined in this article the different criteria to consider and how isahit works with these clients.
Isahit has a wide range of solutions and tools that will help you train your algorithms. Click below to learn more!