
AI & Automation
Augmented Actions, model routing, machine translation, and automation.
27 articles
Augmented Actions Setup
Enable AI and connect your own OpenAI, Azure, or Gemini credentials.
- wxrks Augmented ActionsWhat wxrks Augmented Actions are, how to turn them on at the account, org unit, and Editor level, and how to use each action in the Editor.
- Configuring wxrks Augmented Actions with Azure OpenAI ServiceHow to find the API Key, Resource Name and Deployment Name of your Azure OpenAI resource in the Azure Portal, so wxrks can connect to it for Augmented Actions.
- Configuring wxrks Augmented Actions with Google GeminiStep-by-step guide to connecting Google Gemini (Vertex AI or Gemini Studio) as a custom AI provider for wxrks Augmented Actions, including which one to choose and how to fix a…
- Enabling AI Integration in wxrks through OpenAI or Microsoft AzureEnable Augmented Translation Actions and, optionally, connect your own OpenAI or Azure OpenAI account through Model Management.
- Context Sensitive TranslateHow Context Sensitive Translate turns TM, MT, glossary terms and your Sous-Chef profile into one AI translation, and when a strong TM match wins instead.
- Translation ReasoningTurn on Translation reasoning to see why Context Sensitive Translate made the choices it did, category by category.
- Alternative SuggestionRequest an independent AI rewording of a segment, ignoring TM, glossary, and MT — a genuine second opinion, not a variation on what your TM already suggests.
- ProofreadRun Proofread to have wxrks's AI correct grammar, spelling, and punctuation in a segment's translation, applied immediately and reversible.
- Fix TagsFix Tags asks wxrks's AI to reinsert a segment's source tags into your translation for you, correcting missing, extra, wrong-order, or wrong-position tags in one pass.
- Translation SmellsHow the in-Editor Smells check now uses your project's glossary, translation memory and Sous Chef Instructions — reading the context chips, the TM Match finding, and deciding which warnings to…
- Review AssessmentReview Assessment auto-suggests a change category and comment for your review-step edits, using AI, with an explanation you can always override.
- Auto-Confirming Segments by Confidence ScoreHow to let the Smells confidence score auto-confirm or lock AI-translated segments, how the Confirmed by Confidence Score bucket is priced, and how it shows up as Confidence Score Savings…
- How to lock linguists' Editor settings with shared Editor preferencesCreate shared Editor preferences, lock options on or off for linguists, and choose which preferences each Organizational Unit and project uses.
- How to set up Smells Scoring: severity weights, smell types and confirmation by confidence scoreHow the confidence score is calculated from Smells, and how to set severity weights, per-type severities and the confirmation rule for your account, Organizational Units and projects.
- AI Translation - Output and QualityHow the context you set up for wxrks's AI (Translation Memory, glossary, MT and Sous-Chef profile) shapes translation quality, why the wording can still vary between runs, and how to…
- How to Generate an OpenAI API Key for wxrksStep-by-step instructions for generating an OpenAI API key from your own OpenAI account to use with wxrks Augmented Actions.
Model Routing & MT
How wxrks picks an AI model and configures machine translation.
- Configure Custom AI Providers with Model ManagementPoint wxrks' AI translation features at your own OpenAI, Azure, Anthropic, Groq, OpenRouter, Cerebras, or Google Gemini account, and see exactly how wxrks decides which configured model handles each request.
- How to monitor your account's AI token usageRead the Usage tab in Model Management: token tiles, the trend chart, breakdowns by AI feature and org unit, and why Input plus Output equals Total.
- Connect your own MT engine (bring-your-own MT)Which MT engines you connect yourself, how wxrks connects any other engine, and how it behaves once live.
- Connect your own AI engine for automatic post-editing (Context Sensitive) and quality evaluation (Smells)Route Context Sensitive, Smells and other AI operations to your own engine, and what to send when wxrks connects it for you.
- Connect a Machine Translation (MT) Provider to wxrksConnect and configure MT engines (Amazon, Google, Microsoft, DeepL, ODIN, TransPerfect, Intento), control which one translates each language pair, and see how MT fits into the wxrks translation workflow.
- How routing and locking work in wxrksWhich engine, which people, what moves a segment automatically, and what locks it.
Automation: Agents & Pipelines
Sous Chef agents and Pipelines for automating translation work.
- Sous Chef Agents ExplainedWhat a Sous Chef Agent is, how it shapes wxrks' AI-assisted translations, and when to assign one.
- How to use the Sous Chef profile tab in the Editor
- How to turn your style guides and manuals into a Sous-Chef profile with reference sets
- How to set up Sous Chef Profiles for your account, Organizational Units, and projectsCreate Sous Chef Profiles, set the account default, assign profiles to Organizational Units and projects, map language pairs, and see which profile wins.
- Getting Started with Pipelines in wxrksHow Account Admins set up wxrks Pipelines to automate workflow, task, and project-status changes triggered by account events, including Pipeline Secrets and auto-refreshing OAuth tokens.
