Test before you invest: Fine-tune AI translation before post-editing

No two projects are alike. So why use the same translation setup for each one? 

The right model, the right prompt, and the right amount of context can change from job to job. When you find the best setup before post-editing starts, you get better output and spend less time and money fixing it. 

Where current tools fall short

Most translation management systems (TMS) connect to machine translation (MT) and AI translation engines. Fine-tuning that setup for each project is a different story. It usually falls outside what a TMS can do, or it takes a lot of engineering work. 

What you’ll learn:

• Understand the key differences between MT and AI translation 
• See how PromptLoc, EDIMART’s tool, lets you test and fine-tune AI translation before post-editing 
• Learn how to choose the best model, prompt, and context size for each project 
• Find out how this approach compares to the MT and AI translation built into TMSs 

Who should attend

• Localization and translation managers 
• Project managers at LSPs and enterprises 
• Language engineers and technology leads 
• Anyone looking for more control over AI translation quality 

Bring subject matter experts into memoQ TMS, not your inbox. Join us and learn more about how memoQ’s In-country review tool eliminates unnecessary steps (and thus, room for error) from your translation workflow!

You will listen to

Márton Jánosy – Managing Director, EDIMART

Márton Jánosy is the Managing Director of EDIMART, where he previously worked as a Business Solutions Architect, language engineer and project manager. Besides being involved in the translation industry, he also has a firm background in theoretical linguistics, which he taught at Eötvös Loránd University and the University of Debrecen. Márton is also a self-taught programmer who loves to explore the potential of technology and to innovate.

Balázs Kis – Co-founder, memoQ

Balázs Kis is one of the founders and owners of memoQ. Originally a software engineer, he later earned a PhD in applied linguistics. Since 1994, he has been involved in natural language processing and language technology. Balázs also spent more than ten years in publishing, where he managed titles in computing. He himself was, at one point, the most published author of IT books in Hungarian. At memoQ, he played various roles: he was co-CEO from 2020 to 2024. As memoQ continued to grow, he expanded his professional knowledge in areas like compliance, mediation, and content creation. In 2026, he founded umn content agency, a content agency that focuses on the most valuable content at an organization, created by human hands only.

Share with others