Open Research Position: Data Science / Machine Learning for Design

Location: Cambridge MA (remote option is possible but must have a US work visa/authorization)
Position Type: Part-time, Temporary
Compensation: part-time 10 – 20 hours/week, hourly rate dependent on experience.
Duration: Fall 2024
Job Description:
We are seeking a highly motivated and talented graduate student with expertise in Data Science and Machine Learning to join our research team for a short-term research project. The primary focus of this position is to develop predictive models that can forecast preferences and ratings based on image data, in the context of an on-going design study on the aesthetic perception of wall and floor tiles. This role will provide an excellent opportunity to apply advanced machine learning techniques to a practical problem while contributing to cutting-edge research. This work is part of the Laboratory of Design Technologies and the Material Processes and Systems Group at the Harvard Graduate School of Design.
Key Responsibilities:
• Model Development: Design and implement machine learning models to predict user preferences and ratings from image data.
• Data Processing: Clean, preprocess, and augment mid-size datasets of images to prepare them for model training and evaluation.
• Model Training and Evaluation: Train, validate, and tune models to ensure high accuracy and robustness. Conduct performance evaluations using appropriate metrics.
• Research Documentation: Document methodologies, experiments, and results in a clear and concise manner for both internal use and potential publication.
• Collaboration: Work closely with interdisciplinary team members including designers, architects, industry experts, and supervisors to refine models and achieve research goals.
• Literature Review: Stay updated with recent advancements and trends in machine learning and related fields to incorporate best practices into the project.
Required Qualifications:
• Education: Currently enrolled in, or recent graduate of a graduate program (Master’s or Ph.D.) in Data Science, Computer Science, Machine Learning, or a closely related field.
• Technical Skills:
o Proficiency in machine learning frameworks such as TensorFlow, PyTorch, or Keras.
o Strong programming skills in Python.
• Analytical Skills: Strong understanding of statistical and machine learning algorithms, including supervised and unsupervised techniques.
• Research Experience: Proven experience in conducting research projects, with a strong emphasis on machine learning or data science.
• Problem-Solving: Strong analytical and problem-solving skills with the ability to work independently and in a team environment.

Preferred Qualifications:
• Experience with deep learning frameworks (e.g., TensorFlow, PyTorch).
• Experience with computer vision techniques and libraries such as OpenCV, scikit-image, or similar.
• Publications or significant coursework in machine learning, computer vision, or related areas.
• Familiarity with data visualization tools and techniques.
Application Process:
Interested candidates should submit the following documents:
1. Resume/CV: Detailing relevant educational background and research experience.
2. Statement: Explaining your interest in the position and highlighting any specific expertise that aligns with the job description. Max. 250 words.
3. References: Contact information for at least one academic or professional reference.
Please send your application materials to [email protected] by September 15th 2024.

Robotic Spatial Printing at Rob|Arch 2024

Sulaiman Alothman, Hanspeter Pfister, and Martin Bechthold will be presenting their paper, Toolpath Design Calibration for Robotic Spatial Printing with Supervised Learning in the upcoming Rob | Arch 2024!

“The paper describes a novel method using neural networks to predict and calibrate the toolpath geometries in extrusion 3D printing of natural materials, addressing the challenges of controlling the nonlinear behavior of these materials. Focusing on clay lattice printing, an unorthodox extrusion scenario, the presented method has demonstrated improved accuracy in manufacturing, offering a promising sustainable alternative for contemporary construction.”

Alothman, S.; Pfister, H. & Bechthold, M: “Toolpath Design Calibration for Robotic Spatial Printing with Supervised Learning” in Robotic Fabrication in Architecture, Art and Design (Rob|Arch), 2024.

Future Strategies Keynote

Prof. Martin Bechthold was honored to deliver a keynote lecture at the Future Strategy Symposium, celebrating the 10-year anniversary of the Graduate School of Future Strategy at the Korea Advanced Institute of Science & Technology (KAIST).

Embodied Climates: DDes Conference 23′

Our very own Maroula Zacharias & Katarina Richter-Lunn, along with Malcolm Rio are serving as conference chairs and organizers for this year’s DDes Conference, Embodied Climates. It will be held on September 15 at the Harvard Graduate School of Design. Prof. Martin Becthold will be moderating the Climate Health panel. 

Full conference details here. Livestream link.

Salata Institute Grant: Insulating Buildings with Algae Foam

Exciting News: Together with the Lewis Lab, we are thrilled to announce that we’ve been selected as one of the recipients of the Salata Institute Seed Grant Program!

Our project, Insulating Buildings with Algae Foam will be led by Prof. Martin Bechthold, Prof. Jennifer Lewis, with Dr. Daniel Tish as the postdoctoral fellow dedicated to the research. As reported in the Harvard Gazette: “Most homes in the U.S. are still heated and cooled with fossil fuels, contributing significantly to climate change. Thermal insulation remains the most impactful and cost-effective way to reduce buildings’ energy consumption and emissions — but insulation materials remain too carbon-intensive. Faculty from the Graduate School of Design and John A. Paulson School of Engineering and Applied Sciences will work to develop a proof of concept for a carbon-negative insulation material made from microalgae.” 

Read more on the Harvard Gazette here.

MaP+S Collaboration – La Biennale Architettura 2023

“LAND NARRATIVES | FANTASTIC FUTURES” installation at La Biennale Architettura 2023

MaP+S collaborated as a technical partner for the “LAND NARRATIVES | FANTASTIC FUTURES” installation at La Biennale Architettura 2023, Venice. The installation, led by Toni L. Griffin from Harvard Graduate School of Design and urbanAC founder, showcased the untold stories and creative expressions that emerged despite the segregation and vacant lands in Chicago’s ‘Black Belt’ neighborhoods.

More info : Link

MaP+S Research @ DigitalFUTURES !

MaP+S would like to invite you to join us for an exciting research livestream hosted by Digital Futures !

Whether you are an expert in the field or simply interested in learning more about our research, we would love to have you join us for this engaging and informative event.

The livestream will take place here on 29th April, 10 am EDT.

We hope to see you there! Don’t forget to mark your calendars and spread the word to anyone who may be interested.

Recording to his can be found here : Link

Student Ceramic Projects on Display

During the fall 2022 Nathan King taught the ‘Material Systems: Digital Design and Fabrication’, once again with a focus on ceramic-based design research.  The course continued our decade-long collaboration with the Harvard Ceramics Studio, with support by ASCER Tile of Spain. The show is located in Gallery 224 at the Harvard Ceramics Studio in Allston, and is open to the public from Monday to Friday between 10 am – 4 pm. More information is here.

Exhibition of Student Ceramics Projects

MaP+S Group @ ACADIA 2022

We are excited to share our work at ACADIA 2022: Hybrids & Haecceities in Philadelphia, PA, between Oct 27 -29. We will present a project on clay lattice printing, Responsive Spatial Print Trajectory: 3D Printing of Clay Lattices with Self-Corrective Recalibration, and a paper titled Non-Linear Fabrication: Rapid Data Collection and Toolpath Recalibration in 3D Printing of Clay Lattice. 

We hope to see you there !!!

3d Printing of Ceramic Foams – Advanced Material Technologies

Our collaborative work on 3D printing of ceramic foams has been published in the Advanced Material Technologies journal. The study is the result of our collaboration with the Complex Materials Group led by Prof. Dr. André R. Studart at the ETH in Zurich.

Dutto, A., Zanini, M., Jeoffroy, E., Tervoort, E., Mhatre, S. A., Seibold, Z. B., Bechthold, M., Studart, A. R., 3D Printing of Hierarchical Porous Ceramics for Thermal Insulation and Evaporative Cooling. Adv. Mater. Technol. 2022, 2201109.

Link to the paper here.

ColdSNAP featured in Fast Company

Our work on a novel evaporative cooling system based on our patented ceramic technology was recently featured in Fast Company. Over the summer we were able to install a prototype in Harvard’s HouseZero, gathering data and experience about the real world challenges of cooling with water in Boston’s humid and hot summers. The results look very promising!

Best Paper Award – ICSA 2022

Delighted to announce that our paper titled ‘Post-tensioned Ceramic Structures: Design, Analysis and Prototyping‘ has won the ‘Best Paper Award 2022‘ at the ICSA 2022 5th International Conference on Structures and Architecture , Aalborg, Denmark !!!

The paper can be found here.

MaP+S at ICSA 2022

We are happy to present our work on ‘Post-tensioned Ceramic Structures: Design, Analysis and Prototyping‘ at the ICSA 2022 5th International Conference on Structures and Architecture 06-08 July, Aalborg, Denmark.

The link to the full schedule is here.

The paper can be found here

Katarina Richter-Lunn (MDes ’21) awarded a 2022 Paul & Daisy Soros Fellowship

Courtesy of Katarina Richter-Lunn

Harvard University Graduate School of Design DDes candidate Katarina Richter-Lunn (MDes ’21) has been awarded the 2022 Paul & Daisy Soros Fellowship for New Americans, a merit-based graduate school program for 30 immigrants and children of immigrants. Chosen from a competitive pool of over 1,800 applicants, Richter-Lunn was selected for her potential to make significant contributions through her research and work in the United States. She will join 10 other Harvard affiliated students who are being honored in 2022 by the Soros organization—all of whom receive up to $90,000 to support graduate studies.

More info here.