In 2018, I was independently contracted to edit a manuscript in preparation for submission to Science Robotics by the MIT Mediated Matter Group, led by Neri Oxman. Working directly with the research team across two revision cycles, I provided structural, technical, and copyediting feedback through 57 comments — spanning argument construction, technical rigor, notation consistency, and journal standards — and direct correspondence. Although I declined, the principal investigator invited me to join as a co-author.
Published as: Design of a Multi-agent, Fiber Composite Digital Fabrication System — Kayser, M., Cai, L., Falcone, S., Bader, C., Inglessis, N., Darweesh, B., & Oxman, N. (2018). Science Robotics.
Structural Review
My April 2018 review identified the paper's core problem as architectural, not prose-level: technical content was buried in a catch-all Materials & Methods section, background was undifferentiated from argument, and the Results had no subsection logic connecting design choices to functional requirements. I recommended a near-complete reorganization, with specific content prescribed for each new section.
The structural changes in the submitted draft are closely correlated with those recommendations.
On the Introduction
"It may seem obvious to you, but the introduction is where you draw in laypersons who might not find it so obvious."
Framed the audience problem explicitly — the draft assumed readers already understood the case for autonomous multi-robot construction. The published version opens with extended natural analogies and a direct argument for the system's significance.
On the Robot Criteria section
"I like that you organize this section by design parameters, but it's a little hazy. Introduce them in a clear first paragraph, and then address each one in a distinct paragraph. Perhaps: (1) System-level design parameters (adaptability, tunability, homogeneity, simplicity) followed by individual robot/structure subsystem design parameters: (2) mobility, (3) ground-based vs aerial-based (4) construction material and extrusion/feed processes (5) construction geometry and infrastructure needs... Consider using sub-headers."
The submitted draft reorganized Results into exactly this kind of subsystem structure, with dedicated subheadings for the filament-winding arm, inflatable mandrel, material feed, and drive system.
On Materials & Methods
"All the exciting technical explanations are down here! Consider redistributing to the Results section. Use this section to go into detail that only someone trying to reproduce the robots would need."
Directly implemented: the submitted draft moved all substantive technical description into Results, condensing M&M to hardware specifications and sourcing details.
Section Structure, Before and After
The first draft had five top-level sections with no useful subheadings in Results. The submitted draft has a clearer hierarchy with twelve well-organized subsections — including three comprehensive background sections, a more substantial and logical Results section supported by additional figures and an expanded case study, and two new analytical sections in Discussion.
First draft — April 2018
- Introduction
- Results
- Autonomous Swarm Platform
- Design Workflow
- Architectural-Scale Print
- Discussion
- Conclusion and Considerations
- Future Work
- Materials & Methods
- Composite Material Handling
- Mobility
- Controls & Coordination
- Individual Control
- Multi-robot Coordination
- Human Controller Interface
- Final Structure Fabrication
Submitted draft — July 2018
- Introduction
- Fiber-based composites in architecture
- Construction with multi-robot systems
- Approach
- Results
- Creating Fiberglass Tubes
- Filament-winding arm
- Reusable mandrel
- Material feed
- Drive system
- Programmatic software interface
- Extending to Multiple Robots
- Design for multi-robot fabrication
- Parallel operation
- Case Study
- Discussion
- Robot design impact on output
- Towards full autonomy
- Materials & Methods (condensed)
New or substantially new section
Types of Editorial Contributions
My feedback across both revision cycles spanned four categories, reflecting the range of work required to bring a complex technical manuscript to journal standards.
MIT Graduate Theses
In the same period, I edited the doctoral dissertations and masters theses of candidates at MIT, working in the same mode as the Science Robotics engagement — providing structural, technical, and copyediting feedback directly with the researchers and their advisor.
Scientific Manuscripts
Additional scientific manuscripts edited for MIT researchers.