Staff ML Data Engineer - Multilingual NLP Pipelines

Reflection; Greater London; Sur place; GBP 90 000 - 130 000; Plein temps

Avantages offerts par ce poste

Top-tier compensation

Stock options

Health & wellness

Meals

Life & family

Vacation days

Sponsorship support

Team building

Résumé du poste

Reflection in London is seeking a researcher to design and operate large-scale multilingual data pipelines, including sourcing, cleaning, deduplication, language identification, and script normalization across languages.

You will define quality bars for multilingual corpora, run scientific experiments to scale multilingual models, lead small projects, and collaborate with pre-training and post-training teams to achieve measurable improvements.

Qualifications

• Strong software engineering fundamentals and ability to process web-scale datasets.
• Experience building large-scale data pipelines for language models and MT, speech or search.
• Fluency or working proficiency in at least one language other than English.
• Rigorous, measurement-first mindset: you prove a data change helped.
• Navigate trade-offs between research objectives and engineering realities.
• Comfortable with ambiguity and high ownership in a small, fast-moving team.

Responsabilités

• Design and operate large-scale multilingual data pipelines — sourcing, cleaning, deduplication, language identification, and script normalization across high- and low-resource languages.
• Define and enforce quality bars for multilingual corpora, including translation quality, cultural fidelity, toxicity, and contamination checks.
• Design and run scientific experiments to advance understanding of scaling large language models to improve multilingual data efficiency.
• Lead small research projects independently while collaborating on larger initiatives.
• Build evaluation sets and diagnostics that expose where model behavior degrades by language, register, or domain, and close gaps with targeted data.
• Work with pre-training, mid-training, and post-training teams to land measurable, step-function improvements in multilingual capability.

Connaissances

Software engineering

Data pipelines

Distributed systems

Multilingual NLP

Language proficiency

Measurement-driven

Formation

MS or PhD in Computer Science / ML

Outils

Tokenizers

Crowdsourcing platforms

Description du poste

Reflection in London is seeking a researcher to design and operate large-scale multilingual data pipelines, including sourcing, cleaning, deduplication, language identification, and script normalization across languages.

You will define quality bars for multilingual corpora, run scientific experiments to scale multilingual models, lead small projects, and collaborate with pre-training and post-training teams to achieve measurable improvements.