The Hebrew version is usually the one that waits

The English program is approved and live. Then the request comes: the same training for the team that works in Hebrew. Or it runs the other way, and a program written in Hebrew is needed by a team abroad. Either way it is a second project, and it tends to wait longer than the first one did.

A translation on its own is not the job. Text that reads from the right changes the layout of every screen, from where the navigation sits to which way a progress bar fills. Examples and names that sound natural in one office sound odd in another. A presenter video has to exist in the new language, not just carry subtitles. And someone who speaks the language has to review the result before learners see it, which means finding that person and getting their time.

Katie does this job. She works in more than 20 languages with full right-to-left support, and local teams review each adaptation before rollout. You ask for the version in plain conversation, the way the global rollout demo on her homepage shows, and your team approves it before anyone sees it. If the role is new to you, the guide to the AI instructional designer explains what one does and does not do.

What Katie does with a program that needs another language

She adapts rather than translates

In the homepage demo the request is "We need this training program in Spanish too." Katie's first move is a question: "Latin America or Spain? I'll adapt the language and cultural nuances accordingly." A Hebrew version gets the same treatment. The intent of the original design stays intact, and the wording, the examples and the names are adapted for the people who will use them. The portfolio calls this format localised microlearning: short learning experiences adapted for different languages and working contexts.

She builds for right-to-left reading

Right-to-left support is part of what Katie does, not a fix applied afterwards. The lessons, the practice activities and the assessment are built for a language that reads from the right, so the layout is not a separate job for your team once the words are in place.

She produces the video for that version

Video is on the rate card in three forms: AI presenter video priced by the minute, video with a presenter and voice you select, and an animated explainer with no presenter. Where the original program has a presenter video, the adapted version can have one too, at a known price per minute. Which presenters and voices are available for a given language is a question for the scoping call rather than a promise on this page. Captions, alt text and audio description are included in every training at no credit cost.

A local reviewer approves it

Each adaptation is reviewed by the local team before rollout. That review is one of the approval points, not a courtesy. The reviewer reads it as a learner from that team would, sends feedback in plain words on Teams, Slack or WhatsApp, and Katie makes the changes right away.

She publishes it the same way

The approved version goes to your existing LMS as its own SCORM package, or out as a direct link. The SCORM for your LMS page covers that step.

She keeps the versions current

When a rule changes, you tell Katie what changed. She updates the affected lessons, videos and practice activities and tells you where else the old rule appears. Ask her for the same update in each language you publish, and each version goes back to its reviewer before it goes live.

A worked example: one program, a second language

This is the global rollout request from Katie's own demo on the homepage, with the Hebrew case laid alongside it, so you can see where the path is the same and where it differs.

The program that exists

The team already has the policy training on approval limits from the homepage's policy demo: approved by the L&D manager, published to the LMS, in use.

The request

A manager writes: "We need this training program in Spanish too." For a Hebrew version the request is the same sentence with one word changed. Nobody opens an authoring tool or a translation vendor's portal.

Katie's question

Her reply in the demo is "¡Perfecto! Latin America or Spain? I'll adapt the language and cultural nuances accordingly." The question for a Hebrew version is of the same kind: which team is it for, and which of your own terms should it use. Katie remembers your terminology and business context from project to project, so the answer only has to be given once.

The adaptation

Katie adapts the language and the cultural nuances. The scenarios keep their decisions but get examples that fit the team. The right-to-left version is built as such, with the practice and the assessment laid out for a language that reads from the right. If the original had a presenter video, the adapted version gets its video at the per-minute rate.

The local review

The local team reviews the adaptation before rollout. A reviewer who works in Hebrew every day reads it, sends notes in chat, and Katie shows the new version. The L&D manager approves the complete program.

Publishing

Katie handles the SCORM export. The team packages the Hebrew version for the LMS as its own package, or shares a direct link.

Weeks later, a rule changes

"The approval limit changed this morning. Update it in the policy training." Katie's reply in the demo: "Done! The update is ready for your approval. I also found the old limit in the Manager Onboarding and Procurement courses. Should I prepare updates for those as well?" The same update is prepared for the Hebrew version, reviewed by the same local reviewer, and published when the team decides.

What your team approves, and when

  • The objectives and the structure, before production. For an adaptation, that is the decision about which team it serves and which terms it uses. Nothing is built until you say the plan is right.
  • The complete version, before publishing. The local team reviews each adaptation before rollout, and your team approves it against your own quality bar.
  • Every later change. A rule change, a new language or a department version is prepared for review and published only when you decide.

You own the source materials and every language version of the finished program, including if you stop working with Katie. Before rollout, Trust Katie starts a security review covering your content, learner data, the AI providers involved and your security requirements.

What the learner gets

A program in their own language, built to read from the right, with examples from their own working context rather than a translated version of someone else's. Captions, alt text and audio description come with every training.

Inside the training, learners can talk to Katie. She asks questions, role-plays realistic scenarios and gives feedback grounded only in the material your team approved. In system training, where employees practise a process on screenshot-based screens, Katie is there as their AI guide and answers questions in real time, including in Hebrew.

After launch she analyses learner errors, questions and drop-offs and recommends fixes for you to review and publish, in each version you run.

What it costs in workload

Katie is priced on the workload she takes off your team, never per creator or learner seat. One credit is one cent, and she shows the exact cost before every action.

The lines that matter on this page are the language and video lines. A translation is 500 credits a language, so a Hebrew version of an approved training is a known number before you ask for it. AI presenter video is 400 credits a minute, video with a presenter and voice you select is 800 credits a minute, and an animated explainer with no presenter is 150 credits a video. Captions, alt text and audio description cost no credits, and neither do SCORM export and hosting.

The listed plan is $1,000 a month for 100,000 credits. The pricing page shows that budget as a worked month that includes 10 translations and 40 minutes of presenter video alongside 20 trainings, and it has an estimator you can set to your own languages and volumes. If you would rather prove it first, the pilot is eight weeks with one team, on one real training, from brief to publish.

Or bring the program to a 30-minute scoping call.