Latest AI JobsBrowse Jobs
phrase logo

AI Researcher

phraseCzech RepublicEngineering
Apply on company site
Salary
Not listed
Job type
Full-time
Experience
Entry
Posted
15 Sept 2026
Status
Open · checked 7d ago

About this role

AI Researcher is an open position at Phrase located in the Czech Republic, suited for an early-career professional with strong foundations in machine learning who wants to build production translation models and agentic workflows. Phrase builds a Language Intelligence Platform combining AI, agentic orchestration, and API-first architecture.

Role Overview

As an AI Researcher at Phrase, the selected individual joins a team building and evaluating proprietary translation models and agentic workflows for live production systems. This position involves running experiments, fine-tuning models, and constructing evaluation pipelines alongside senior researchers. The role focuses on applied natural language processing and machine translation within an international environment.

Key Responsibilities

Fine-tune and evaluate large language models for translation and language-quality tasks using methods like LoRA.
Build evaluation pipelines for translation quality utilizing automatic metrics such as BLEU, TER, ChrF, MQM, and COMET, alongside LLM-as-judge methods.
Construct and evaluate agentic workflows featuring multi-step reasoning, tool use, and structured output generation.
Implement components within the team model-development program under a defined scope.
Run reproducible experiments and document results clearly.
Cooperate with Product, Engineering, and Solutions teams to turn research into shippable capabilities.
Benchmark frontier and open-source models against internal baselines.
Support quality-evaluation systems, style-guide integration, data pipelines, and feedback loops for continuous model improvement.

Requirements

Solid grounding in machine learning and natural language processing with hands-on experience fine-tuning or working with large language models or transformer-based models.
Working knowledge of machine translation, machine translation evaluation, information extraction, text generation, or search.
Exposure to or strong interest in agentic or large language model-orchestrated systems.
Experience running experiments and applying automatic and human-in-the-loop evaluation metrics.
Solid Python skills with machine learning and natural language processing libraries such as Hugging Face or PyTorch.
Master of Science in Computer Science, Computational Linguistics, or a related field, or a Bachelor of Science with relevant experience, projects, or a strong portfolio.
Coachable attitude with a habit of asking questions, acting on feedback, and sharing results early.

Nice to Have

A PhD in a related field.
Internships, thesis work, or projects in natural language processing or machine translation with production or applied exposure.
Familiarity with localization or machine translation concepts like MQM, COMET, post-editing, or translation memory system integrations.
Exposure to experiment tracking tools like Weights & Biases or agentic-pipeline frameworks such as PydanticAI or LangGraph.
Publications or open-source contributions in natural language processing topics.

Skills and Tools

Python
Hugging Face
PyTorch
Weights & Biases
PydanticAI
LangGraph

Salary and Benefits

Four additional company holidays per year
A Christmas break without using vacation allocation
Birthday off
Two giveback days for community support and volunteering
Professional and extensive onboarding
Enterprise Claude License
Additional local benefits depending on the hiring entity

About the Company

Phrase provides a Language Intelligence Platform that combines artificial intelligence, agentic orchestration, and an API-first architecture. The company operates globally with offices and remote team members across Europe, the UK, the US, and the Asia-Pacific region, fostering an environment based on collaboration, innovation, and shared purpose.

How to Apply

Apply on the official company careers page.

Skills and tech stack

PythonHugging FacePyTorchWeights & BiasesPydanticAILangGraphLLMLLMsNLPMachine LearningFine-tuning

Questions candidates ask

What experience level is required for this role?

The position is built for someone earlier in their research career with strong fundamentals who wants to grow their depth in applied natural language processing.

What educational background is expected?

An MSc in Computer Science, Computational Linguistics, or a related field is required, or alternatively a BSc with relevant experience, projects, or a strong portfolio.

Where is the job located?

The job location is the Czech Republic, within a company that has a global team based in offices and remote colleagues across Europe, the UK, the US, and the APAC region.

What specific technical libraries and tools are used?

The tech stack includes Python, Hugging Face, PyTorch, Weights & Biases, PydanticAI, and LangGraph.

Are there any specific time-off benefits mentioned?

Yes, benefits include four extra company holidays per year, a Christmas break, the employee's birthday off, and two giveback days for volunteering.

Ready to apply?

The application is on the official phrase careers page.

Apply on company site