Thesis: Human-in-the-Loop Sanitization for Hybrid Local and Cloud LLM Agents under Data Sovereignty Constraints (AI)
Контакты
Формат работы
onsite
Тип работы
fulltime
Грейд
trainee
Английский
c1
Страна
Sweden
aillmopen sourcecloud computinghuman in the loopagentic aidata sovereigntydata sanitization
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
84
Хорошая вакансия
развернуть
У дипломного проекта понятная исследовательская задача, практическая разработка и план оценки, но это временный студенческий проект, а не постоянная должность.
?Только для выпускников
Чёткая тема проекта
Передовая тема ИИ
?Работа в офисе
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Thesis: Human-in-the-Loop Sanitization for Hybrid Local and Cloud LLM Agents under Data Sovereignty Constraints (AI): Building and evaluating an escalation path from a local coding agent to a cloud LLM through human-approved sanitization, with an accent on sensitive-data removal, reversible context mapping, and provider-agnostic agent harnesses. Focus on comparing rule-based and LLM-based sanitizers, measuring coding-task performance and review effort, and assessing setup complexity and local hardware requirements.
Location: Onsite at the hirify.global office in Sundsvall, Sweden
Company
hirify.global is a strategic partner in product and digital service development, formed by the combination of Semcon and Knightec.
What you will do
Investigate and build an escalation path from a local coding agent to a cloud LLM within an agentic coding harness.
Develop a sanitization step that removes or generalizes sensitive identifiers before a user approves an outgoing request.
Compare one-way sanitization with reversible sanitization that restores original names and context in returned answers.
Keep the solution provider-agnostic, using OpenCode or another suitable agent harness.
Evaluate leakage, coding-task performance, human review effort, setup complexity, suitable anonymization tools, and local hardware requirements.
Use synthetic or open-source code with seeded sensitive identifiers; do not use real sensitive data.
Requirements
Bachelor’s or Master’s degree in Computer Science or Computer Engineering.
Expected graduation in June 2027.
Fluency in Swedish, both written and spoken.
Fluency in English, both written and spoken.
Nice to have
Experience or interest in local and cloud LLMs, coding agents, agentic harnesses, anonymization, or data sovereignty.
Culture & Benefits
Hands-on thesis work on real projects in a supportive and innovative environment.
Work from the Sundsvall office as part of hirify.global’s product and digital service development organization.
Start in January or March 2027.
Classified information, legal compliance assessment, and cryptographic approaches such as homomorphic encryption are out of scope.
Hiring process
Submit the application through the careers page only; applications are accepted until 30 November 2026.
Applications should be submitted as soon as possible. Questions may be sent to johanna.edstrom@knightec.se.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →