Brahma AI is the end-to-end content system for content creation. We build photorealistic digital humans with our proprietary models, ATMAN (visual identity) and VAANI (voice), which power an enterprise platform that lets these humans speak across languages. We are now building a dedicated in-house linguistics function so that our output is not just accurately translated, but linguistically and vocally excellent in every language we ship.

About The Role

This is not a traditional translation seat. You will be one of our first linguists in, owning language quality across both text and voice, and helping shape the tools that produce it and how other linguists interact with these tools. You will sit close to the VAANI (voice) team. Your job is twofold:

• Make our output linguistically excellent.
• Make the systems that generate and support that output better, across both text and voice.
  
  

You will be the first internal user of our localisation tooling and voice interfaces. Your feedback will directly define how they work.

Today we rely on external vendors and internal native speakers. That does not scale, and it does not give us ownership. We need dedicated linguists who hold the quality bar and drive the tooling forward. You will, however, be involved in supervising external linguists as volume demands.

What You Will Do

• Own linguistic quality (review, QC, correction) for translated text in your language(s).
• Review and evaluate synthesised voice from VAANI in your language, starting with linguistic correctness, then naturalness, prosody, pronunciation, emotion, and timing.
• Work directly with the VAANI team to define and improve how voice is generated and evaluated in your language.
• Help design the linguist-facing platform experience: the workflows, review screens, and tooling linguists use to translate and validate output. You are the first user, so you tell us what good looks like.
• Build and maintain the linguistic assets for your language: style guides, glossaries, evaluation criteria, pronunciation guidance, and a pronunciation dictionary you contribute to and continually refine.
• Set the quality standards and turnaround expectations for external vendors and internal speakers, rather than absorbing all the volume yourself.
• Turn linguistic problems into structured, actionable input for engineering and product.
  
  

Who You Are

• An excellent linguist first. Native or near-native command of your language, with strong English.
• Formal training or equivalent professional experience in linguistics, translation, or localisation.
• Fluent with AI. You already use AI tools in your work and you trust them where they earn it. You want to help shape the models and platform tools that produce the highest quality translation and voice output, in the most efficient way.
• Adaptable and hands-on. Comfortable in a fast-moving, undefined environment. You like building process where none exists.
• A clear communicator. You can explain linguistic nuance to engineers and turn vague product goals into concrete language requirements.
• Quality-obsessed but pragmatic. You hold a high bar and you ship.
  
  

Nice To Have

• Experience with voice, TTS, or speech evaluation.
• Experience with localisation tooling, CAT tools, or QA platforms.
• Enough technical literacy to understand, at a high level, how these models behave and fail.
• Additional languages.
  
  

Languages We Are Hiring For

We are hiring linguists across the following languages:

• English
• Mandarin Chinese (Simplified)
• Spanish (Latin American and Castilian)
• Hindi
• Arabic (MSA)
• Portuguese (Brazilian)
• Japanese
• Korean
• French
• German
• Italian
• Turkish
  
  

Why Join?

You will help define how digital humans speak across the world’s languages, with direct ownership of both the quality and the tools that produce it. This is a build role, not a maintenance role.

Please mention the word **EASE** and tag RMjYwMDoxOTAwOjA6MmQwMDo6NjAx when applying to show you read the job post completely (#RMjYwMDoxOTAwOjA6MmQwMDo6NjAx). This is a beta feature to avoid spam applicants. Companies can search these words to find applicants that read this and see they're human.