Skip to content
View almondshawarma's full-sized avatar

Highlights

  • Pro

Block or report almondshawarma

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
almondshawarma/README.md

Howdy, I'm Aman 👋

I'm an undergraduate nuclear engineering student at Texas A&M, currently doing precision-measurement physics at the Cyclotron Institute! I really like understanding things from first principles, building the instrument, deriving the math, then finding where it connects to everything else.

My long-term goal is working on magnetic confinement for fusion energy, but for now I'm lucky enough to work on some genuinely beautiful physics up close, and it's humbling how far first principles get you.


🔬 What I'm working on now

  • TAMUTRAP: Research on a cylindrical Penning trap measuring β–ν angular correlations in nuclear beta decay, searching for physics beyond the Standard Model.
  • LSTAR: I built the control system for the Cyclotron Institute's isobar separator, multipole ion optics driving a high-voltage crate over SNMP, cleaning the ion beam to a mass resolving power of M/ΔM ≥ 3000!

🛠️ Featured projects

TAMUTRAP-timing: The control software for TAMUTRAP's pulsed timing chain. It turns high-level pulse descriptions into low-level SpinCore PulseBlaster instructions so the trap catches ions at the right microsecond. One pure-Python core drives both a Tkinter operator GUI and a headless EPICS IOC. Every mode can run dry, with or without the hardware present. Python · SpinAPI · caproto (EPICS) · Tkinter

LSTAR-MPOD-control: The production control software for LSTAR. It computes electrode voltages from physics parameters (quadrupole / hexapole / octupole strengths), driving a WIENER MPOD crate over SNMP, with three-layer safety validation, audit logging, and both GUI and CLI frontends. Python · puresnmp · NumPy · Tkinter

Helios: A self-hosted platform that organizes my own life, whether documents, health, library, finances, or notes, and exposes it to AI through a privacy-tiered layer. It has ~25 Docker services on one machine, with a Model Context Protocol layer enforcing which data can and can't reach a cloud model. It also runs Hyperion, a little toy "curiosity engine" that surfaces research, a "money lens" over public science funding. Docker · Python · Model Context Protocol · Ollama · Grafana · Tailscale

🧰 Instruments I'm building

These are all "different" projects, but I find that I keep bumping into the same signal chain (and its awesome!) of sensors, conditioning, digitization, filtering, feature extraction, and then state estimation. Each of these is honestly kind of just an excuse to learn a little more of that chain end-to-end, finding where theory and measurement disagree in various parts of my life.

  • Car DAQ: turning my Honda Civic into an open-hardware mobile instrument, a Raspberry Pi 5 reading classic-CAN, GPS, and an IMU, logging vehicle dynamics, road roughness, and the RF environment on every drive. A phone-based pipeline (phyphox + dashcam) already logs real drives today, and there's a time-synced trip viewer (map + telemetry + dashcam video). A build guide, and BOM, are in the repo, so hopefully anyone can recreate it!
  • Ionosphere DAQ: (meant to be) a citizen-science ionospheric sensor node, receive a satellite beacon with a single antenna and an RTL-SDR, and pull two real numbers out of one pass: a Doppler-curve fit (observed drift vs. TLE prediction) and an amplitude-fading period. In all honestly, this is pretty preliminary- one antenna only gives you amplitude and frequency, so it can't really hand you a TEC number just yet. Phase II (a coherent dual-channel receiver) is where a real Faraday-rotation TEC measurement becomes possible. Runs on radio, a reproducible headless SDR host for a Raspberry Pi.
  • Hestia: home automation spoken for by EPICS, the same control system that runs particle accelerators, light sources, and ITER. Every light, sensor, and blind in my apartment will eventually become a process variable by the time I'mdone, so I get to learn the real accelerator/fusion control stack (PV naming, Channel Access, Phoebus, the archiver) by operating one at apartment scale (and teaching my roommates some cool stuff along the way!).
  • Hall-effect magnetometer: built from first principles, Biot–Savart, vector calculus, calibration. (Repo coming soon)
  • Guitar acoustic-resonance mapper: Helmholtz resonance and the wave equation on a bounded domain, via FFT. (Repo coming soon)

✍️ I write things down

I keep a public digital garden! These are working notes and half-formed ideas across physics, math, and engineering, from tacos to Penning traps.

🧪 Tools I've been using

Python Linux Docker Git LaTeX NumPy Arduino

🔗 Elsewhere

Website LinkedIn


Aman's GitHub stats Top languages

Pinned Loading

  1. LSTAR-MPOD-control LSTAR-MPOD-control Public

    Production SNMP control system for the LSTAR isobar separator's multipole electrodes (TAMU Cyclotron Institute/TAMUTRAP), with a CLI + GUI, shared physics core, and layered hardware-safety checks.

    Python 6

  2. TAMUTRAP-timing TAMUTRAP-timing Public

    Production control system for TAMUTRAP's pulsed timing chain (TAMU Cyclotron Institute), programming a SpinCore PulseBlaster from a shared pulse-math core, with a tkinter GUI and an EPICS IOC, all …

    Python 1

  3. hestia hestia Public

    A college apartment project run on beamline/reactor controls! Home-built ESP32 IoT controlled with EPICS.

    Python 2