Research Scientist, Foundation Model

Prior Labs

WHO WE ARE Foundation models transformed text and images. Structured data - the largest and most consequential data format in the world - stayed untouched. Tables run every clinical trial, every financial model, every scientific experiment, every business decision, and no one had built a foundation model that truly understood them. Until now. What LLMs did for language, we're doing for tables. The next modality shift in AI is happening, and we're hiring the team that makes it. Momentum. We pioneered tabular foundation models and are now the world-leading organization in structured-data ML. Our TabPFN v2 model was published as a Nature cover story and set a new state of the art for tabular machine learning. Since release we've scaled model capabilities 20x+, passed 3.5M+ downloads and 7,500+ GitHub stars, and are seeing accelerating adoption across research and industry - from detecting lung disease with Oxford Cancer Analytics to preventing train failures with Hitachi to improving clinical-trial decisions with BostonGene. The hardest work is ahead. We're scaling tabular foundation models to millions of rows, thousands of features, real-time inference, and entirely new data modalities, while building the infrastructure to run them in production across some of the most demanding industries on earth. These are open problems no one else is working on at this level. Our team. We're a small, highly selective team of 30+ engineers, researchers, and GTM specialists, with backgrounds spanning Google, Apple, Amazon, DeepMind, Meta, Microsoft Research, G-Research, Jane Street, Goldman Sachs, and CERN. We're led by Frank Hutter, Noah Hollmann, and Sauraj Gambhir, and advised by world-leading AI researchers including Bernhard Schölkopf and Turing Award winner Yann LeCun. We ship fast, do top-tier research, and hold each other to an extremely high bar. What's next. In 2025 we raised €9m pre-seed led by Balderton Capital, backed by leaders from Hugging Face, DeepMind, and Black Forest Labs. The next phase of growth is here, which makes this an ideal time to join. CORE AREAS OF IMPACT You'll be among the first scientists collaborating and working an entirely new class of AI models, not just incremental improvements. As an early-stage startup working on foundation models for tabular data, we have countless exciting research ideas and problems to explore - you're sure to find challenges that match your interests and expertise. We are working on problems such as: * Scaling our transformer architectures from 10K to 1M+ samples while maintaining performance * Building multimodal models that combine text and tabular understanding on proprietary data * Developing specialized architectures for time series, forecasting, and anomaly detection * Creating efficient inference methods for production deployment * Researching causal understanding in foundation models * Designing novel approaches for handling multiple related tables WHAT WE'RE LOOKING FOR * Currently pursuing or holding a PhD in Computer Science, Applied Mathematics, Statistics, Electrical Engineering, or a related field (we will also consider exceptional Master's students) * Deep experience with ML frameworks, especially PyTorch and scikit-learn * Strong engineering fundamentals with excellent Python expertise * Experience in data-science and working with tabular data or time series * Publications at top-tier venues (NeurIPS, ICML, ICLR) or significant open-source contributions BENEFITS * Strong mentorship and professional development opportunities * Work with state-of-the-art ML architecture, substantial compute resources, and a world-class team * Comprehensive benefits including healthcare, transportation, and fitness Life at Prior Labs We're a small, ambitious team solving one of the hardest problems in AI, and we're just getting started. You'll work closely with world-class researchers and builders who care deeply about the quality of their craft, the impact of their work, and the people they work with. We move fast, we think rigorously, and we take the time to do things right. If you're excited by hard problems, motivated by real-world impact, and want to be part of building something that matters, we'd love to hear from you. We're building our teams in Berlin, Freiburg, and New York and we believe that when you're working on something as hard and exciting as TabPFN, being in the same room matters. Most of our roles are based in one of our offices but great people come from everywhere, and in exceptional cases we're open to remote. This usually involves frequent travel to one of our offices and the whole company comes together regularly for offsites to think, build, and celebrate together. Our Commitments We believe the best products and teams come from a wide range of perspectives, experiences, and backgrounds. That's why we welcome applications from people of all identities and walks of life, especially anyone who's ever felt discouraged by "not checking every box." We're committed to creating a safe, inclusive environment and providing equal opportunities regardless of gender, sexual orientation, origin, disability, or any other trait that makes you who you are. We care about how your data is handled. Read our Recruiting Privacy Notice to see exactly what we collect, why, and how long we keep it.

Prior Labs

WHO WE ARE Foundation models transformed text and images. Structured data - the largest and most consequential data format in the world - stayed untouched. Tables run every clinical trial, every financial model, every scientific experiment, every business decision, and no one had built a foundation model that truly understood them. Until now. What LLMs did for language, we're doing for tables. The next modality shift in AI is happening, and we're hiring the team that makes it. Momentum. We pioneered tabular foundation models and are now the world-leading organization in structured-data ML. Our TabPFN v2 model was published as a Nature cover story and set a new state of the art for tabular machine learning. Since release we've scaled model capabilities 20x+, passed 3.5M+ downloads and 7,500+ GitHub stars, and are seeing accelerating adoption across research and industry - from detecting lung disease with Oxford Cancer Analytics to preventing train failures with Hitachi to improving clinical-trial decisions with BostonGene. The hardest work is ahead. We're scaling tabular foundation models to millions of rows, thousands of features, real-time inference, and entirely new data modalities, while building the infrastructure to run them in production across some of the most demanding industries on earth. These are open problems no one else is working on at this level. Our team. We're a small, highly selective team of 30+ engineers, researchers, and GTM specialists, with backgrounds spanning Google, Apple, Amazon, DeepMind, Meta, Microsoft Research, G-Research, Jane Street, Goldman Sachs, and CERN. We're led by Frank Hutter, Noah Hollmann, and Sauraj Gambhir, and advised by world-leading AI researchers including Bernhard Schölkopf and Turing Award winner Yann LeCun. We ship fast, do top-tier research, and hold each other to an extremely high bar. What's next. In 2025 we raised €9m pre-seed led by Balderton Capital, backed by leaders from Hugging Face, DeepMind, and Black Forest Labs. The next phase of growth is here, which makes this an ideal time to join. What you'll do This role is foundational data science: building the foundations of tabular foundation models so a single model can solve data-science problems across the board. Roughly half the work is inventing new frontier tools for TFMs, and half is building the dataset and benchmark bedrock they stand on. * Invent and build the frontier tools that extend TabPFN, including its thinking, scaling, and agentic capabilities, and the new methods that let one model generalize across the full landscape of data-science problems. This is the most open-ended part of the work and grows over time. * Set the research direction by deciding which model capabilities and benchmarks are worth pursuing, choosing what is worth solving rather than optimizing a score someone else set. * Bring in external research and real customer needs to shape new model and tooling directions, and publish frontier results that move the field forward. * Build trustworthy benchmarks from the structured data behind real, high-impact problems, so the team optimizes for real-world performance rather than one leaderboard. * Faithfully implement the baselines and competitor models that set the gold standard of applied data science, giving the team a read on where TabPFN leads and where there is room to improve. * Build an automated, agentic pipeline with a human in the loop so this data and benchmark foundation scales to far larger volumes without losing rigor, itself a genuinely new tool. What we're looking for * You have solved data-science problems across many domains and datasets to a high standard, optimizing for strong performance across a whole suite of tasks rather than the single best score on one. * You work undogmatically across the ML toolbox, including getting strong results with gradient-boosted trees (such as XGBoost) and not only with deep learning. * You understand the common categories of dataset defects (leakage, label noise, distribution shift, duplication, mislabeled targets, and similar) and why each corrupts a training or benchmark signal. * You are energized by foundational work, valuing the dataset and benchmark bedrock as much as the frontier tooling, and you have taken on hard problems others passed over. * You thrive as a senior individual contributor in an ambiguous, early-stage, low-process environment. You are opinionated on best practice in Data Science and can make good judgement calls on approaches to complex problems. Nice to have * Experience building or extending evaluation harnesses, benchmark suites, or experiment frameworks that others rely on. * Experience building LLM- or agent-assisted pipelines with a human in the loop to scale a previously manual workflow. * Experience acting as the link between external research or customer needs and an internal model or product roadmap. * Prior work on tabular, structured-data, or foundation-model problems, or helping shape an emerging research subfield through community work. Life at Prior Labs We're a small, ambitious team solving one of the hardest problems in AI, and we're just getting started. You'll work closely with world-class researchers and builders who care deeply about the quality of their craft, the impact of their work, and the people they work with. We move fast, we think rigorously, and we take the time to do things right. If you're excited by hard problems, motivated by real-world impact, and want to be part of building something that matters, we'd love to hear from you. We're building our teams in Berlin, Freiburg, and New York and we believe that when you're working on something as hard and exciting as TabPFN, being in the same room matters. Most of our roles are based in one of our offices but great people come from everywhere, and in exceptional cases we're open to remote. This usually involves frequent travel to one of our offices and the whole company comes together regularly for offsites to think, build, and celebrate together. Our Commitments We believe the best products and teams come from a wide range of perspectives, experiences, and backgrounds. That's why we welcome applications from people of all identities and walks of life, especially anyone who's ever felt discouraged by "not checking every box." We're committed to creating a safe, inclusive environment and providing equal opportunities regardless of gender, sexual orientation, origin, disability, or any other trait that makes you who you are. We care about how your data is handled. Read our Recruiting Privacy Notice to see exactly what we collect, why, and how long we keep it.

Krafton

우리는 게이머의 로망을 실현하기 위해, 누구도 가지 않는 길을 갑니다. 예상을 뛰어넘는 과감한 상상력과 기술로, 전 세계 팬들이 잊지 못할 세상을 만들기 위해 담대하게 도전하고 개척합니다.   We pioneer the path to players' dreams. With bold imagination and breakthrough technology, we create unforgettable worlds for fans across the globe. ----------------------------------------------------------------------------------------------------------------------------------   ※ 본 공고는 2026년도 전문연구요원 현역 신규편입 대상입니다. 조건 부적합 시 불합격으로 안내드리는 점 참고 부탁드립니다. AI RESEARCH 본부에 대한 더 많은 내용은 크래프톤 AI RESEARCH 본부 홈페이지 (아래 이미지 클릭) 및 크래프톤 AI RESEARCH 본부 소개 영상을 통해 확인하실 수 있습니다. (바로가기 : KRAFTON AI)   우리 팀(프로젝트)을 소개합니다.  지난 2021년부터 크래프톤은 우리가 가진 게임 제작역량을 새롭게 확장할 수 있는 영역이자 새로운 도전을 지속할 분야로 딥러닝을 제시하고 해당 분야 연구에 집중하고 있으며, 본격화를 위하여 핵심 인재 영입 등 적극적인 투자를 진행하고 있습니다. [AI Research 본부 비전] 크래프톤 AI Research 본부는 사내외 여러 분야와 협업 하여 다양한 문제에 대한 AI 솔루션을 제공하며, 자체 딥러닝 기술 연구를 통해 우리만의 서비스를 개발합니다. 그 방향성은 크게 두가지입니다. * 게임 제작에 도움이 될 수 있는 딥러닝 기술 개발을 통한 게임 제작 효율성 및 보편성 증대 * 유저와 함께 게임 플레이를 할 수 있는 Multi-modal LLM 기반 Foundation Model 개발 [Foundation Model R&D] Foundation Model Research 팀은 AI Research 본부의 비전을 바탕으로 MLLM 핵심 기술을 연구·개발하여, 새로운 플레이 경험을 만드는 데 기여합니다. * 텍스트, 이미지, 오디오, 게임 로그 등 다양한 데이터를 다루는 파운데이션 모델을 설계·구축합니다. * 대규모 분산 학습 환경에서 모델 아키텍처, 학습 알고리즘, 데이터 파이프라인, 평가 전 과정에 대해 이론적·실무적 통찰을 축적합니다. [Culture Fit] AI Research 본부 인원들은 다양한 프로젝트를 통하여 여러 분야의 팀원들과 교류하고 협업하며 다양한 문제에 창의적인 아이디어를 제시할 수 있습니다. 나이 및 직급에 관계없이 자유롭게 의견을 개진하는 분위기가 장려됩니다. 다양한 문화적 배경을 가진 인원들이 모여 팀을 이루고 있으며 활발한 소통을 위해 통·번역 등 언어적 장벽을 해소할 수 있는 방법들을 적극 지원하고 있습니다.   우리 팀과 함께할 미션을 소개합니다. * 대규모 Multi-modal LLM 학습 및 선행 연구를 진행합니다. * 모델의 성능 향상을 위한 Modeling / Data / Evaluation pipeline을 고도화 합니다. * 연구 성과를 Tech Report 혹은 NeurIPS, ICLR, ICML 등 탑티어 학회에 논문·워크숍 등의 형태로 발표합니다.   이런 경험을 가진 분과 함께 성장하고 싶습니다! (필수요건) * 2026년도 내 전문연구요원 현역 신규편입으로 입사 가능하신 분 (석사 이상 학위를 취득 또는 석/박사 통합과정 수료 대상) * 딥러닝 관련 분야 전공 석/박사 졸업 이상 혹은 이에 준하는 연구 경력 *  AI/ML 탑티어 학회 (NeurIPS, ICLR, ICML 등) 관련 분야 탑티어 저널 논문 작성을 경험하신 분 * 연구를 주도할 수 있으며 딥러닝 관련 논문을 통해 빠르게 이해하고 실험 설계할 수 있는 능력 * 새로운 domain에 빠르게 적응할 수 있는 분 * 원활한 업무 커뮤니케이션 능력 * 해외 출장에 결격 사유가 없는 분   이런 경험들이 있다면 저희가 찾는 그 분입니다! (우대요건) * AI/ML 탑티어 학회 (NeurIPS, ICLR, ICML 등) 관련 분야 탑티어 학회 발표/수상 경험이 있으신분 * 20B 이상 대규모 Foundation Model에 대한 학습/배포 경험이 있으신분   ---------------------------------------------------------------------------------------------------------------------------------- 크래프톤의 도전에 함께 하기 위해 아래의 전형 과정이 필요합니다. * 서류 전형 > 사전 인터뷰 (Phone Interview) > *직무 테스트 (Pre-Test) > 직무 면접 (Technical Fit Interview) > 종합 면접 (Culture Fit Interview) > 학/경력, 평판조회 > 처우협의 > 최종 합격 및 입사 *해당 전형은 포지션에 따라 변동될 수 있습니다. * 각 전형의 세부 진행 방식은 개별 안내드립니다. * 필요 시 직무 테스트 또는 면접 전형이 추가될 수 있으며, 이에 대한 상세 내용은 개별 안내드립니다. * 전형 결과는 지원서에 기재된 이메일로 2주 이내 발송되며, 내부 일정에 따라 다소 지연될 경우 별도 안내드립니다. * 본 공고는 상시채용으로 진행되며, 우수 인재 채용 시 조기 마감될 수 있습니다.   필요 서류를 확인해주세요! * 입사지원서 (CV 형식/가이드라인), 자기소개서, 경력기술서, 포트폴리오 (필수) * 신입일 경우 자기소개서를, 경력일 경우 경력기술서를 중심으로 기술해 주시기 바랍니다. * 포트폴리오 첨부 시, 하단 안내 사항을 확인해 주시기 바랍니다.    근무지 * 역삼 센터필드 West 타워   고용형태  * 정규직 (단, 후보자 처우 협의 결과에 따라 고용형태가 변경될 수 있음)     ---------------------------------------------------------------------------------------------------------------------------------- 아래 안내 사항을 확인해주세요! * 장애인 및 국가 유공자 등 취업 보호 대상자는 관계 법령에 따라 우대합니다. * 지원서 내용 중 허위사실이 있는 경우에는 합격이 취소될 수 있습니다. * 5개월의 수습기간을 적용합니다. 회사는 수습기간에 대한 평가 결과에 따라 본채용을 거부할 수 있으며, 수습기간 중이라도 중간평가 결과에 따라 수습기간을 조기 종료하고 본채용을 거부할 수 있습니다. 수습기간 내 고용형태 및 급여 조정은 없습니다.  * 채용 전형 중 궁금하신 사항은 크래프톤 채용 FAQ 내에서 확인하실 수 있습니다.   BE BOLD, LEARN AND WIN! 크래프톤의 성장과 도전의 스토리를 알아가고 싶다면?  * 크래프톤 제작 스튜디오 * 크래프톤 게임  * 크래프톤 비전 & 핵심가치 * 크래프톤 라이프 & 복리후생 * 크래프톤 공식 유튜브 채널 * 배틀그라운드 공식 유튜브 채널 * 크래프톤 인스타그램 * 크래프톤 블로그 ---------------------------------------------------------------------------------------------------------------------------------- PLEASE CHECK THE INFORMATION BELOW * Those eligible for an affirmative action program such as individuals with disabilities or with distinguished service to the state will be given hiring preference according to relevant laws * False statements in your resume may lead to the withdrawal of the employment offer * A 5-month probationary period applies. Based on the mid-term evaluation, employment may be terminated early or not continued. No changes in employment type or salary during this period. * Please contact career@krafton.com for other inquiries BE BOLD, LEARN AND WIN! WOULD YOU LIKE TO EXPLORE KRAFTON’S GROWTH AND ADVENTURE STORIES? * KRAFTON Production Studios * KRAFTON Games * KRAFTON Vision and Core Values * KRAFTON Life & Welfare Benefits * KRAFTON Youtube Channel * BATTLEGROUNDS Youtube Channel * KRAFTON Instagram * KRAFTON Blog