Senior AI Researcher<\/p>\n\n\n\n
At Phrase, we help open the door to global business by providing the world’s leading Language Intelligence Platform.<\/p>\n\n\n\n
The Phrase Platform combines AI, agentic orchestration, and a headless, API-first architecture in one composable system. Beyond translation, it orchestrates and adapts content to culture, audience, channel, brand voice, and intended outcome in any language, for every audience. It applies the context that makes content perform in every market: quality standards, glossaries, prior translations, and cultural nuance. Every team, in every region, can ship content that is on-brand, on-point, and ready for any audience.<\/p>\n\n\n\n
Phrase gives enterprises the intelligence to automate workflows, the freedom to connect their own tools and engines, and the control to govern global content at scale.<\/p>\n\n\n\n
With a global team based in our offices, and remote colleagues across Europe, the UK, the US, and the APAC region, Phrase offers an international environment built on collaboration, innovation, and shared purpose.<\/p>\n\n\n\n
Phrase’s AI Research team builds, trains, and evaluates the proprietary translation models and agentic workflows that power the platform’s language quality \u2014 live systems serving production traffic, not proof-of-concept work. This role takes ownership of a significant research workstream within that portfolio, extending Phrase’s live systems and shaping the direction of the team’s wider research. You’ll work across fine-tuning and instruction-tuning LLMs for domain-specific machine translation, designing evaluation frameworks that combine automatic metrics and LLM-as-judge approaches, and building agentic workflows for multi-step reasoning and tool use. It’s built for someone with strong hands-on research experience who wants real scope to lead a body of work, mentor others, and grow their technical influence over time.<\/p>\n\n\n\n
What you’ll be responsible for:<\/p>\n\n\n\n
\u2022 Design, train, and evaluate LLMs for translation and language-quality tasks, using fine-tuning techniques such as LoRA and DPO and instruction-tuned models.<\/p>\n\n\n\n
\u2022 Design and implement evaluation frameworks for translation quality, combining automatic metrics (BLEU, TER, ChrF, MQM, COMET), LLM-as-judge approaches, and hybrid pipelines.<\/p>\n\n\n\n
\u2022 Build and evaluate agentic workflows for NLP problems, spanning multi-step reasoning, tool use, and structured output generation.<\/p>\n\n\n\n
\u2022 Own a significant workstream within the team’s model-development programme, from data and training through evaluation and production integration.<\/p>\n\n\n\n
\u2022 Design and run reproducible experiments, documenting results and decisions clearly.<\/p>\n\n\n\n
\u2022 Work with Product, Engineering, and Solutions to turn research findings into shippable capabilities.<\/p>\n\n\n\n
\u2022 Benchmark frontier models \u2014 OpenAI, Gemini, Claude, and open-source \u2014 against internal baselines.<\/p>\n\n\n\n
\u2022 Evolve the team’s quality-evaluation systems, including style-guide integration and customer-led, outcome-oriented metrics.<\/p>\n\n\n\n
\u2022 Shape decisions on active learning, feedback-loop design, and data-pipeline strategy for continuous model improvement.<\/p>\n\n\n\n
\u2022 Mentor junior researchers, review PRs and research write-ups, and contribute to reading groups and deep dives.<\/p>\n\n\n\n
\u2022 Contribute to hiring, including take-home review and technical interviews.<\/p>\n\n\n\n
\u2022 Represent your work in cross-functional forums with Product and Engineering.<\/p>\n\n\n\n
What you need:<\/p>\n\n\n\n
\u2022 Strong hands-on experience fine-tuning LLMs or large transformer-based models; familiarity with pre-training and preference-tuning (RLHF\/DPO) is desirable.<\/p>\n\n\n\n
\u2022 Solid NLP background, with demonstrable expertise in at least one of: machine translation, MT evaluation, information extraction, text generation, search optimization, or reinforcement learning.<\/p>\n\n\n\n
\u2022 Ability to design, build, and evaluate agentic systems, including tool-calling agents, multi-agent pipelines, or LLM-orchestrated workflows.<\/p>\n\n\n\n
\u2022 Experience designing and running evaluation frameworks, including automatic metrics and human-in-the-loop evaluation design.<\/p>\n\n\n\n
\u2022 Strong Python skills and experience with ML\/NLP libraries such as Hugging Face or PyTorch.<\/p>\n\n\n\n
\u2022 Familiarity with workflow orchestration tools such as Flyte, Prefect, or Airflow (desirable).<\/p>\n\n\n\n
\u2022 PhD in Computer Science, Computational Linguistics, or a related field, or an MSc with equivalent hands-on research or industry experience.<\/p>\n\n\n\n
\u2022 Industry experience with production NLP systems, including shipping models, handling data pipelines, and monitoring quality in live traffic (desirable).<\/p>\n\n\n\n
\u2022 Familiarity with the localization and MT domain, including MQM, COMET, post-editing workflows, segment-level quality signals, or TMS integrations (desirable).<\/p>\n\n\n\n
\u2022 Experience with experiment tracking tools such as Weights & Biases, and model-serving or cloud ML infrastructure such as vLLM, TGI, AWS SageMaker, or GCP Vertex AI (desirable).<\/p>\n\n\n\n
\u2022 Experience with frameworks for structured agentic pipelines such as PydanticAI or LangGraph (desirable).<\/p>\n\n\n\n
\u2022 Strong publication record or open-source contributions in NLP\/ML (desirable).<\/p>\n\n\n\n
\u2022 Product-minded, thinking about research in terms of what it enables and who it serves, not just whether the numbers went up.<\/p>\n\n\n\n
\u2022 A natural driver who identifies open problems and moves work forward without waiting to be pushed.<\/p>\n\n\n\n
\u2022 Growth-oriented, motivated to expand technical scope and mentor others, with a path toward greater research leadership over time.<\/p>\n\n\n\n
Our current tech stack in this role:<\/p>\n\n\n\n
\u2022 Python<\/p>\n\n\n\n
\u2022 Hugging Face<\/p>\n\n\n\n
\u2022 PyTorch<\/p>\n\n\n\n
\u2022 Weights & Biases<\/p>\n\n\n\n
\u2022 PydanticAI<\/p>\n\n\n\n
\u2022 LangGraph<\/p>\n\n\n\n
\u2022 vLLM \/ TGI<\/p>\n\n\n\n
\u2022 AWS SageMaker \/ GCP Vertex AI<\/p>\n\n\n\n
\u2022 Flyte \/ Prefect \/ Airflow<\/p>\n\n\n\n
What you’ll get:<\/p>\n\n\n\n
– Work experience in a successful and growing global SaaS company<\/p>\n\n\n\n
– Be part of an international team in Europe, APAC, and the Americas<\/p>\n\n\n\n
– Expert colleagues in their field who are determined to build the best localization platform on the market, creating a world where language never limits opportunity<\/p>\n\n\n\n
– An agile work environment, where it is encouraged to take smart risks<\/p>\n\n\n\n
– Take part in a culture full of trust, support and loyalty, where respectful and open feedback is valued, and diversity is fully embraced<\/p>\n\n\n\n
– A positive, open-minded, and innovative atmosphere<\/p>\n\n\n\n
– Support in your professional development and personal career goals<\/p>\n\n\n\n
What’s on top:<\/p>\n\n\n\n
– 4 Company holidays additional to your regular holidays (1 day per quarter where the entire company is off to celebrate our achievements).<\/p>\n\n\n\n
– In addition, the company also provides employees with a Christmas break to allow you to spend time with family and friends without use of your vacation allocation.<\/p>\n\n\n\n
– Your birthday is off because it is important to celebrate you as well.<\/p>\n\n\n\n
– 2 Giveback days where you can support the local community, volunteer, and\/or participate in charity events and activities.<\/p>\n\n\n\n
– Professional and extensive onboarding.<\/p>\n\n\n\n
– Enterprise Claude Licence<\/p>\n\n\n\n
– Additional local benefits depending on the entity you’re hired at, just ask your Talent Acquisition Partner<\/p>\n\n\n\n
Phrase is committed to ensuring equal pay for equal work and work of equal value between women and men. Our aim is that our workforce will be truly representative of all sections of society and that each worker feels respected and able to give their best. The job requirements and salary range for this job have been established through a gender-neutral job evaluation and classification process. This process objectively assesses the skills, responsibility, effort and working conditions required for each job. This systematic approach ensures fair pay regardless of who performs the work. We welcome applications from all qualified candidates regardless of their sex, race or ethnicity, disability, religion\/belief, sexual orientation, gender identity or age. We value and welcome different perspectives, experiences and backgrounds as we believe that these differences make our team even stronger on our mission of opening the door to global business by giving everybody access to the content they need in the language they speak.<\/p>\n","protected":false},"template":"","meta":{"_acf_changed":false,"_stopmodifiedupdate":false,"_modified_date":"","_searchwp_excluded":"","episode_type":"","audio_file":"","podmotor_file_id":"","podmotor_episode_id":"","cover_image":"","cover_image_id":"","duration":"","filesize":"","filesize_raw":"","date_recorded":"","explicit":"","block":"","itunes_episode_number":"","itunes_title":"","itunes_season_number":"","itunes_episode_type":"","footnotes":""},"phrase-job-locations":[2788],"phrase-job-departments":[2705],"class_list":["post-154458","phrase_job","type-phrase_job","status-publish","hentry"],"acf":false,"taxonomies":[{"term_id":2788,"name":"Czech Republic","slug":"czech-republic","term_group":0,"term_taxonomy_id":2788,"taxonomy":"phrase_job_location","description":"","parent":0,"count":4,"filter":"raw"},{"term_id":2705,"name":"Engineering","slug":"engineering","term_group":0,"term_taxonomy_id":2705,"taxonomy":"phrase_job_department","description":"","parent":0,"count":2,"filter":"raw"}],"_links":{"self":[{"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-jobs\/154458","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-jobs"}],"about":[{"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/types\/phrase_job"}],"version-history":[{"count":2,"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-jobs\/154458\/revisions"}],"predecessor-version":[{"id":155119,"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-jobs\/154458\/revisions\/155119"}],"wp:attachment":[{"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/media?parent=154458"}],"wp:term":[{"taxonomy":"phrase_job_location","embeddable":true,"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-job-locations?post=154458"},{"taxonomy":"phrase_job_department","embeddable":true,"href":"https:\/\/phrase.com\/es\/wp-json\/wp\/v2\/phrase-job-departments?post=154458"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}