CONSULTANCY - Linguistic technology support Consultant -SSA-2026-HRCS-01

LSP completed a machine translation (MT) retraining in 2024-2025 for its customized models in five languages. These models require regular retraining based on new texts translated since the last time. The download of bitexts in the five languages required for the training, as well as their cleaning are extremely time-consuming tasks and are done mainly manually. LSP seeks to automatize these processes through the development of a specific tool that would download bitexts from LSP CATT tool (eLUNa) and clean them for training purposes. The models will need to be retrained as well.

• Develop a tool (macro, script, etc.) with customized filters to download automatically WMO bitexts from eLUNa, WMO computer-translation assisted tool, in priority for the years 2024-2026, but not limited to, for the following language pairs: English > Arabic, Chinese, French, Russian and Spanish.
• Create a database of downloaded bitexts used for MT training in all the six languages
• Develop a tool (macro, script, etc.) to clean up the downloaded bitexts in the five language pairs by removing duplicates, same language matches, empty targets, misalignments, etc.
• Retrain Microsoft Translator models per above language pairs (English > Arabic, Chinese, French, Russian and Spanish) and train the reversed models (Arabic, Chinese, French, Russian and Spanish > English)
 

•Evaluate the retraining results based on automatic scoring (BLEU/COMET) for the retrained Microsoft models and provide a report on scoring results to LSP
• In cooperation with WMO linguists, complete human evaluation of the retrained Microsoft models for the five language pairs listed above
• Provide a report on human evaluation of retrained Microsoft models
• In cooperation with Linguistic Services Support Officer, prepare bilingual glossaries based on WMO terminology and acronyms, clean them for training and integration into Microsoft Translator models
• Retrain all Microsoft Translator models using the prepared glossaries for each language pair
• Evaluate the results of retraining based on the integration of glossaries for each language pair and provide a report on retraining outcome
• If the retraining results are not satisfactory, take action to improve the existing engines to the extent possible, or conduct research and recommend an alternative solution
• Train the Linguistic Services Support Officer and any relevant staff member, as deemed necessary by LSP, on retraining models and provide guidance material for future retraining and engine maintenance
• Present the results of retraining to LSP stakeholders
• Document the procedure for retraining and evaluation of models
 

• Provide support in testing functionalities of on-going technology projects and advise on their acceptability or need for further elaboration.
• Train the Linguistic Services Support Officer (LSSO) and any other stakeholders, as deemed necessary by LSP, on the use and maintenance of the language tools (macro, script, flow and other related components).
• Prepare detailed documentation and other guidance material describing the relevant processes to be provided to the LSSO
• Carry out other relevant duties as required

• Assist LSP by developing language tools related to translation, editing and publishing processes such as, but not limited to: cleanup of Word files before upload in the CATT tool in the six official languages; validation of references; validation of hyperlinks, alignment of glossaries, etc.
• Amend WMO template macro (“WMO Ribbon”) to connect with Sharepoint lists and session websites and other enhancements, as required
 

• Knowledge of training and retraining of machine translation models (Microsoft Translator required, other engines a distinct advantage)
• Advanced knowledge of Microsoft applications (Azure, Sharepoint, Forms, Power Automate, VBA, etc.);
• Expertise in Python and similar programming language
• Ability to transfer knowledge by documenting processes and providing training

University degree in information technology, data science, computer science, language technology or related field.

Proven experience in:
- Training and retraining of machine translation models (Microsoft and other engines)
- Developing language tools (via macros, scripts, flows, etc.) using Microsoft applications and IT scripting tools such as Python, as necessary
- Developing and managing databases/records sets

Excellent knowledge of English (both oral and written). Knowledge of other official languages of WMO would be an advantage.
(Note: The official languages of WMO are Arabic, Chinese, English, French, Russian and Spanish.)

Pay Band C : (USD 390.- / USD 560.-)

One year full time

The consultancy can be carried out remotely, but the consultant will be required to come to WMO Secretariat in Geneva for in-person meetings as deemed necessary by the hiring manager. WMO will not cover any travel expenses; they are expected to be fully paid by the consultant

Sarah Eymann (seymann@wmo.int)

Applications should be made online through the WMO e-recruitment system.

Do not send your application via multiple routes. WMO no longer accepts applications via post or email. Only applicants for whom WMO has a further interest will be contacted. Shortlisted candidates may be required to sit a written test and/or an interview.

WMO is committed to achieving diversity and a balanced workforce. Applications are welcome from qualified women and men, including those with disabilities. The statutory retirement age is 65. Pursuant to WMO Standing Instructions, the minimum age to be eligible for consideration for vacant positions is 18, and the maximum age must enable the candidate to serve for at least the term of the contract before reaching mandatory age of separation.

WMO does not tolerate harassment, sexual harassment, exploitation, discrimination and abuse of authority. All selected candidates, therefore, undergo relevant checks and are expected to adhere to the respective standards and principles.

WMO does not charge a processing fee at any stage of its recruitment, selection, and hiring processes (i.e., application stage, interview stage, validation stage, or appointment and training). WMO will not ask for applicants’ bank account information.