We are a small London startup with the ambition to change education with ML-powered tutoring. Our flagship product is a mobile application for teaching English to intermediate and advanced learners.

We’re on the verge of solving one of the biggest challenges in education – making high-quality, personalised learning accessible to everyone. We are building a fundamental model for education - one that can accurately predict student knowledge and orchestrate lessons, adapting to the students needs.

We’re looking for a Senior NLP Engineer, with a proven track record of delivering ML models to production, to join the ML team in our growing company.

What you will do 🚀

• Build fundamental models for education - solving the ultimate learning task of predicting student knowledge and optimal ‘next task’
• Work with a vast amount of unique data - we have data from over 1M language tests, including text and voice data
• Create brand new dictionaries and train models to determine the difficulty of words, idioms, phrasal verbs etc.
• Analyse large amounts of diverse data - including data from every movie, book, and song
• Work in a cross-functional team and communicate with backend engineers and product managers
• Create new types of tests for language learners to gather more test results, analyse them, and build prediction models based on these results
• Optimise and fine-tune machine learning models for performance, scalability, and accuracy
• Building fully automated pipeline for dictionary building; including span identification, word sense distribution, and sense granularity decision.

Essential skills 🙏

• Strong expertise in NLP
• Complete end-to-end experience - from finding and cleaning data all the way to monitoring models in production
• Strong understanding of neural networks, CNNs, RNNs, LSTMs, and transformers
• Experience building automated data pipelines
• Hands-on experience with LLM tooling and libraries (e.g., Hugging Face Transformers/PEFT, tokenisers, spaCy or similar)
• Experience shipping NLP systems: prompt engineering, fine-tuning (e.g., LoRA/PEFT), vector search, and RAG-based services.
• Great knowledge of NLP algorithms: tokenisation, embeddings, attention, language modelling, text classification/generation, and information retrieval