Hey there! I am currently pursuing a PhD at the University of Cambridge, focusing on Machine Unlearning.

I previously worked in the Tech & Digital Advantage team of the Boston Consulting Group (BCG) and co-founded a deep tech startup funded by the European Space Agency Business Incubator (ESA BIC).

Cam logoETH logobcg logoesa biclogo ibmresearch logo

Academic Experience

Research
Google Scholar

  1. 2nd PhD Year

    Potion: Towards Poison Unlearning
    Schoepf S., Foster J. & Brintrup A.
    Journal of Data-Centric Machine Learning Research (DMLR)

    Loss-Free machine Unlearning
    Foster J.*, Schoepf S.* & Brintrup A.
    ICLR 2024 Tiny Papers | * equal contribution

    Parameter-tuning-free data entry error unlearning with adaptive selective synaptic dampening
    Schoepf S., Foster J. & Brintrup A.
    Preprint (under review)

    Zero-Shot Machine Unlearning at Scale via Lipschitz Regularization
    Foster J., Fogarty K., Schoepf S., Öztireli C & Brintrup A.
    Preprint (under review)

  1. 1st PhD Year

    Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening
    Foster J.*, Schoepf S.* & Brintrup A.
    AAAI 2024 | * equal contribution

    Identifying contributors to manufacturing outcomes in a multi-echelon setting: a decentralised uncertainty quantification approach
    Schoepf S., Foster J., & Brintrup A.
    IEEE Transactions on Industrial Informatics

    AgentChat: Multi-Agent Collaborative Logistics for Carbon Reduction
    Xu L., Mak S., Schoepf S., Ostroumov M. & Brintrup A.
    Preprint (under review)

    Implementation of Autonomous Supply Chains for Digital Twinning: a Multi-Agent Approach
    Xu L., Proselkov Y., Schoepf S., Minarsch D., Minaricova M. & Brintrup A.
    IFAC World Congress 2023

    Data augmentation and synthetic data generation for low-frequency and sparse data problems
    One of two co-facilitators for a team of 11 researchers
    The Alan Turing Institute Data Study Group

  1. Before PhD

    Using Reinforcement Learning for the Three-Dimensional Loading Capacitated Vehicle Routing Problem
    Schoepf S., Mak S., Senoner J., Xu L., Netland T. & Brintrup A.
    IJCAI 2023 Workshop, Search and Planning with Complex Objectives

    Pattern Recognition ANNs: Automotive Brake Creep Groan as Case Study
    Pürscher M., Schoepf S. (oral presentation), & Fischer P.
    8th Congress of the AAAA (2018)

Education

  1. University of Cambridge, Clare College (UK)
    PhD in Engineering (Oct. 2022 - Present)
    ‍    - ‍Funded by UKRI EPSRC DTP
  2. ETH Zürich (CH)
    ‍    MSc in Management, Technology, and Economics (Sep. 2019 – Oct. 2021)
    ‍    ‍- Won the semester-long Marketing Analytics prediction competition (data from a Swiss FI; offered fast track recruitment)‍
    - 6-Month thesis on RL for combinatorial optimization (IJCAI workshop paper)
  3. Graz University of Technology (AUT)
    BSc in Mechanical Engineering and Business Economics (Sep. 2015 – Dec. 2018)
    ‍    ‍- Thesis and paper (Alps Adria Acoustics Association Congress) on brake-creep-groan detection via artificial neural networks
    - Circle of Excellence 2018 (one-year program for ∼ 20 students per year from all universities in Graz; 50.000+ students)

Working Experience

  1. IBM Research (IE)
    ‍    AI Security Research Intern (June 2024 – Sep 2024)
    ‍    ‍- Foundation model security research (Automated Red Teaming) in the AI Security & Privacy team
    - Best presentation award at the IBM Research International Intern Presentation Day (UKI)
  2. Boston Consulting Group AG (CH)
    ‍    Associate (Nov 2021 – Sep 2022)
    ‍    - Performed a vendor due diligence for one of three use cases of a cybersecurity asset (market model, KPCs, etc.)
    - Part of the SAP x BCG ”tech-enabled sustainability transformation” Moonshot PMO team
    ‍- Created a digital maturity assessment framework for a leading financial institution at process step granularity level to estimate savings potential in USD, acting as the information backbone for a multi-year digital transformation
  3. Audili OG - European Space Agency Business Incubator (AUT)
    ‍    Co-Founder (Dec 2018 – April 2021)
    ‍    ‍- High-resolution soil type prediction via spectral satellite imagery
    - Space Oscar winner at Copernicus Masters 2019, Europe’s largest earth observation competition initiated by ESA
    - Designed a Kedro framework with MlFlow, boto3, Optuna, custom sklearn pipelines and custom loss functions
    - Successfully drafted business plans/applications for the European Space Agency Business Incubator, Copernicus Accelerator, Clinton Global Initiative University, and Parsec 100 (Horizon 2020), totalling over €60.000 equity-free funding
  4. Mercedes-Benz AG (DE)
    ‍    Business Development Internship at Future Transportation (May 2019 – Aug. 2019)
    ‍    ‍- OSS Onboard Logic Unit platform
    - Used the engineering concept of morphological boxes to create a modular framework for possible go-to-market strategies
    - Drafted and evaluated the data monetisation and value creation parts of the business model
  5. OMV AG & Institute of Innovation and Industrial Management (AUT)
    ‍    Project Manager (Oct. 2017 – May 2018)
    ‍    ‍- Corrosion detection in pipelines of the national oil and gas company OMV AG
    - Designed an IoT solution estimated by OMV AG to result in 75% savings in monitoring interventions
    - Led an interdisciplinary team of five students from ideation to PoC, reporting directly to the upstream technology manager
  6. Austrian Service Abroad (USA)
    ‍    Translator and Tour Guide; Virginia Holocaust Museum (Jul. 2014 – Jul. 2015)
    ‍    ‍- Selected as one of the top 30 applicants to represent Austria abroad (mandatory Austrian civil service)
  7. ‍‍Add. Internships: Quality assurance for automotive die-casting products; injection molding for motorcycle parts; etc.

Extracurricular Experience

  1. Sigma Squared Society (formerly Kairos Society)
    ‍    Fellow (May 2020 – Present)
    ‍    ‍- Sigma Squared Society is a non-profit organisation and global community of the world’s most ambitious young founders selected for their impact-driven initiatives and their ability to transform broken industries. (prev. Kairos Society)
  2. Analytics Club at ETH Zürich
    ‍    Board Member (Sep. 2019 – Mar. 2021)
    ‍    ‍- Led the DS meetups series (see Analytics Club at ETH YouTube)
  3. TUG Racing – Formula Student
    Data Acquisition Team Member (Oct. 2018 – Apr. 2019)
    ‍    ‍- Modeled and implemented a limited-slip differential for the lap time simulation
  4. icons – consulting by students
    Board Member and Student Consultant (Nov. 2016 – Oct. 2018)
    ‍    ‍- Led 20+ student consultants at the Graz office of Austria’s largest student consultancy
  5. Add. memberships: Tera Eco Racing Team, Austrian Service Abroad State Coordinator, Traditional Music Band, Austrian Water Rescue
  6. Hackathons: 1st at the Green Tech Jam hackathon (behavior nudging), Hackathon for the blind (supermarket navigation), etc.
LinkedinGitHubGoogle Scholar Downhill Unicycling Email: ss2823[at]cam.ac.uk