Eder Medina

CV

email: e_medina [at] g [dot] harvard [dot] edu
alt: em2368 [at] princeton [dot] edu

Research Interests

My attention is currently focused on solving problems within the realms of machine learning, embodied intelligence, and the application of artificial intelligence to engineering and manufacturing. My life goal is to design complex engineering systems capable of evolution and adaptation. Proficient in optimization techniques and manufacturing, I bring expertise to the intersection of cutting-edge technology and practical engineering solutions.

White Papers

• Arcadia: Autonomous manufacturing systems. (Current focus)
  Reach out for more information.
• Exomorphic: White paper on high pressure die casting automation.
  Development on going.
• Maiz.dev: Two year exploratory venture on applying machine learning to machining.
  Decision: pivot.

Education

• Postdoctoral Research Associate, Princeton, 2024
• Postdoctoral Fellow, Harvard, 2022
• Engineering Science PhD, Harvard, 2022
• Engineering Science SM, Harvard, 2020
• Mechanical Engineering BS, UT-Austin, 2016

Publications

• Real-time design of architectural structures with differentiable mechanics and neural networks
  
• Automated discovery of reprogrammable nonlinear dynamic metamaterials
  
• 3D spatiotemporally scalable in vivo neural probes based on fluorinated elastomers
  
• Nonlinear shape optimization of flexible mechanical metamaterials
  
• Neuromechanical Autoencoders
  
• Harnessing Mechanical Deformation to Reduce Spherical Aberration in Soft Lenses
  
• Navigating the landscape of nonlinear mechanical metamaterials for advanced programmability
  

Misc.

One of my life goals is to build a robot that can "do a kickflip!"
I just hope no one beats me to it. If you can fund this, I will drop everything I am currently working on.
Hint: sim2real + RL.

• WREI: Wristwatch Reverse Engineering Intelligence ideation.
• SPILF List: Side projects I'd like to finish (SPILFS). A little slice of what I think about.
• Automatic Differentiation for frequency sweep analysis (2014): Or How Ignorance, Grit, and a Bit of Chain Rule Changed Everything.

Vertical integration hides behind the claim of control and complexity—but AI is shifting the rules. Reach out to discuss how intelligent systems can unlock repairability, restore transparency, and reimagine access at every level. Or even challenge the premise itself: do we, as a society, still need or want this control back? I'm still trying to figure this out.