Mark Prainito

Harvard College, Class of 2029. Joint concentration in Chemistry & Physics and Engineering Sciences (Electrical and Computer Engineering), with a secondary in Mathematics.

I study physical reservoir computing: whether chemical, biological, and photonic systems can be made to compute, and how well. On the hardware side I'm building a hemoglobin-based optical computer, microfluidic devices, CubeSat structures, and a robotic microscope.

Download CV (PDF) Email me LinkedIn
Σ input x(t) physical reservoir fixed, random coupling readout y(t) Win Wout trained dashed: target
How a physical reservoir computer works: an input signal x(t) is coupled into a physical system with rich, fixed dynamics, the reservoir. Nothing inside it is trained. Only the readout learns: a weighted sum of the reservoir's states that recovers the target y(t). My research asks which physical systems do this well, and why.

About

I'm an A.B. candidate at Harvard College, expected 2029, with a joint concentration in Chemistry & Physics and Engineering Sciences (Electrical and Computer Engineering) and a secondary in Mathematics. I grew up in Bellmore, New York.

I split my time between two labs. In the Creative Nanobiophotonics Lab I design nanolaser systems for microscopy and photonic computing, and I build the models behind physical reservoir computing: eight physics- and chemistry-informed reservoir architectures, with Python, MATLAB, and COMSOL workflows that have run billions of simulations across time-series and image-classification benchmarks. I'm first author on a manuscript in preparation, The Physics of a Good Chemical Reservoir, and I'm developing ICHOR, a hemoglobin-based computing platform for in situ biological monitoring, as its next step. In the Weitz Lab I make microfluidic emulsions and work on micro-NMR: measuring droplets in a microcoil, in flow, with an eye on the chemistry of serpentinization.

Outside the lab I design CubeSat hardware with Harvard SEDS, model rocket engines with the Harvard Rocket Propulsion Group, direct outreach for the Harvard-MIT Mathematics Tournament, and run Empowered Innovations, a nonprofit I founded that designs and 3D-prints assistive devices for people with disabilities.

I also play piano, guitar, and drums, play hockey, run, scuba dive (which started with marine-biology research), hike, and fish.

Areas of interest

  • Reservoir Computing
  • NMR
  • Aerospace Systems
  • Electrical Engineering
  • Sensors & Signal Processing
  • RF Design
  • Robotics & Microrobotics
  • Photonics & Photonic Integrated Circuits
  • Lasers
  • Nanotechnology
Mark Prainito standing beside his research poster, The Physics of a Good Chemical Reservoir, at a symposium at Rice University.
Presenting The Physics of a Good Chemical Reservoir at a research symposium at Rice University, Houston, summer 2026.

Research and engineering

What I'm working on, and where.

  1. Harvard University and Rice University. PI: Prof. Sangyeon (Fred) Cho

    • Design nanolaser systems for advanced microscopy and photonic computing, and experimentally validate a light-controlled chemical reservoir that shows nonlinear dynamics and fading memory.
    • Developed eight physics- and chemistry-informed reservoir-computing architectures, with scalable Python, MATLAB, and COMSOL workflows for billions of simulations across time-series and image-classification benchmarks.
    • Leading a first-author manuscript on the physics of chemical reservoirs, and developing ICHOR, a hemoglobin-based computing platform for in situ biological monitoring, as its next step.
    • Reservoir computing
    • Nanophotonics
    • Python
    • MATLAB
    • COMSOL
    Lorenz attractor, projected on x and z closed loop begins x(t) time true Lorenz system reservoir forecast, fed its own output the two agree for several Lyapunov times, then chaos wins
    From my work in the Cho Lab: a chemical reservoir, a nonlinear reaction network driven by light, is trained to forecast the Lorenz system. In closed loop the readout’s own prediction is fed back as the next input, and a good reservoir tracks the true trajectory for several Lyapunov times before chaos pulls the two apart. Which physical properties make a reservoir good is the subject of my manuscript in preparation.

  2. Harvard University. PI: Prof. David A. Weitz

    • Make and characterize microfluidic emulsions: droplets generated on chip, with controlled size and composition, as the samples for NMR measurements.
    • Micro-NMR: NMR on nanoliter-to-microliter samples, where a microcoil wound around the channel gives the sensitivity that a tiny volume otherwise lacks and lets droplets be measured in flow, in situ, without taking them out of the chip. I support the instrumentation and the quantitative analysis of the spectra.
    • Chemistry of interest: serpentinization, the hydration of olivine-rich rock that releases hydrogen and drives hydrothermal chemistry, from methane formation to the energy sources of deep-subsurface life. Microfluidics and NMR together offer a way to follow reactions like these in small, well-controlled volumes.
    • Microfluidics
    • Micro-NMR
    • Emulsions
    • Serpentinization
    • Soft matter
    NS B0 microcoil, RF out and in oil droplets in water, flowing droplets through a microcoil, in B₀ z ∥ B₀ 90° RF pulse M Larmor precession, ω₀ = γB₀ decaying with T₂; the coil picks up the signal 90° acquire free induction decay time Fourier transform ↓ H₂O, 4.7 ppm oil, 1.3 ppm chemical shift
    Micro-NMR on a chip: oil droplets in water flow through a microcoil wound around a microfluidic channel, inside a strong static field B₀. A 90° radio-frequency pulse tips the nuclear magnetization into the transverse plane, where it precesses at the Larmor frequency and decays with T₂; the same coil picks up the induced signal, the free induction decay, and a Fourier transform turns it into a spectrum. Water and oil protons sit at different chemical shifts, so the peak areas read out the composition of the emulsion. In the Weitz Lab I make the emulsions and work on these microfluidic NMR measurements.

  3. Harvard College Students for the Exploration and Development of Space

    • Designed and manufactured CubeSat mechanical hardware with CAD and structural and thermal analysis; coordinated electrical and software integration for an April 2026 SpaceX Falcon 9 launch.
    • SolidWorks
    • Structural analysis
    • Thermal analysis
    • Systems integration

  4. Harvard University

    • Develop analytical and computational models of rocket-engine performance, propellant flow, and subsystem integration.
    • Propulsion modeling
    • Fluid dynamics
    fuel oxidizer valves chamber M < 1 throat, M = 1 exit, M > 1 injector feeds the chamber; the nozzle expands the gas F = ṁ·ve + (pe − pa)·Ae thrust = momentum flux + pressure term gravity turn altitude time Δv = ve · ln(m0 / m1), the rocket equation
    Rocket propulsion in one picture: propellants flow from the fuel and oxidizer tanks through the injector into the combustion chamber; the nozzle chokes the flow at the throat (M = 1) and expands it supersonically, and thrust is the exhaust momentum flux plus the pressure imbalance at the exit. On the launch side the rocket equation, Δv = ve ln(m0/m1), says how far exhaust speed and mass ratio can take you. With the Harvard Rocket Propulsion Group I develop analytical and computational models of engine performance and propellant flow.

Designs and projects

Hardware, CAD, and software I've designed and built. Each one has its own page with drawings and details.

Teaching and activities

  1. Harvard University

    • Selected as the sole course assistant for a cross-listed undergraduate and graduate laboratory electronics course; support instruction, grading, and debugging in analog and digital circuit design.
    • Analog electronics
    • Digital electronics
    • Instrumentation
    input: 100 Hz signal + kHz noise + − R1 Rf R C non-inverting op-amp stage gain = 1 + Rf / R1 = 11 first-order RC low-pass fc = 1 / 2πRC ≈ 1 kHz output: ×11, noise filtered fc signal noise, −14 dB gain vs frequency
    A bench classic from Physics 123: a non-inverting op-amp stage sets the gain (1 + Rf/R1 = 11) and a first-order RC low-pass with fc = 1/2πRC strips the high-frequency noise, so a small, noisy signal comes out clean and eleven times larger. The frequency response shows why: the 100 Hz signal sits on the flat part of the curve while the kilohertz noise is pushed down the −20 dB per decade slope. I’m the course assistant for Physics 123/223, where students build and debug circuits like this on the bench.

Activities

  • Harvard-MIT Mathematics Tournament, Outreach Director

    Since Sept 2025. Harvard-side outreach, partnerships, communications, and event logistics for programs serving 1,000+ participants, including collaborations with competitions at Stanford, Princeton, Yale, and Carnegie Mellon.

  • Empowered Innovations Inc., Founder and Co-President

    Since Jan 2024. A nonprofit that designs and 3D-prints assistive devices for people with disabilities.

  • Tutoring and research mentoring

    Competition math from AMC 10 through AIME and USAMO, with practice sessions built on my own problem sets, plus mentoring for research competitions such as Regeneron STS and ISEF.

Honors

  • Regeneron Science Talent Search Scholar

    2025

  • Eagle Scout

    Boy Scouts of America

CV and skills

Download CV (PDF) Open in browser

Updated October 2026. Education, research, teaching, leadership, honors, and skills on one page.

Programming and HDL
  • Python
  • MATLAB
  • C++
  • Verilog
Engineering and modeling
  • SolidWorks
  • Altium
  • COMSOL
  • Analog and digital circuits
  • Oscilloscopes
  • Structural and thermal analysis
Fabrication and cleanroom
  • Photolithography
  • Electron-beam lithography
  • Thin-film deposition
  • Reactive-ion etching
  • Wet etching
  • Lift-off
  • Spin coating
  • SEM
  • NMR characterization
  • Microfluidics
  • CNC machining
  • 3D printing
Selected coursework
  • Laboratory Electronics
  • Quantum Mechanics
  • Statistical Mechanics
  • Electrodynamics
  • Probability
  • Linear Algebra
  • Multivariable Calculus
  • Organic Chemistry
  • Microfabrication*
  • Robotics*
  • CS50*
  • Spacecraft Engineering*
  • Inorganic Chemistry*

* enrolled for spring 2027

Let's connect

I'm always interested in discussing reservoir computing, microfluidics and NMR, hardware and instrumentation, and collaborative research or engineering opportunities.

Email me

© 2026 Mark Prainito. All rights reserved.