M.K.

Melani Kirstein

I build at the intersection of med-tech, research, and content — less portfolio, more field notes from someone who reads signal traces for fun.

This page is M.K.: my story, the nerdy metadata, and how I actually operate. Jump below if you only have thirty seconds.

// edit this block with your real story when ready

draft.voice

I think in systems, prototype in code, and explain in public — med-tech because the stakes are real, content because teaching is how I learn, research because rigor is the only moat that compounds.

Chapters

The arc

Five beats — scroll through, or jump here if you already know the headline.

01 — Origin

Where curiosity met constraint

I didn't set out to work in med-tech. I set out to understand systems — biological, technical, organizational — and kept finding myself at the intersection where those systems fail people.

Early on, I learned that the gap between what medicine promises and what infrastructure delivers is not a knowledge problem. It's a build problem.

02 — Going Deep

Technical foundations

I went deep on the hard parts: signal processing, data pipelines, validation workflows, the unglamorous infrastructure that makes clinical tools trustworthy.

Code was never the destination. It was the instrument — a way to test hypotheses, ship prototypes, and learn faster than reading alone allows.

03 — Building in the Open

Operator mode

Entrepreneurship, for me, is not about startups as identity. It's about seeing a process that's painful when it doesn't have to be, and choosing to fix it.

I've worked across stealth builds, open-source tools, and research collaborations — always with the same filter: does this make the next person’s work easier or harder?

04 — Med-Tech Lens

Why this domain

Med-tech sits at the edge of what's possible and what's regulated, what's humane and what's scalable. That tension is where I do my best work.

I'm drawn to problems where a wrong abstraction costs more than a delayed launch — where rigor isn't optional and creativity isn't decorative.

05 — Now

What I'm building toward

Today I'm focused on research-grade tooling, clinical signal infrastructure, and the connective tissue between lab bench and real-world deployment.

This site is where I document that work — not as a personal diary, but as field notes for anyone building in the same direction.

Resume

Research record

Patents and publications from Google Scholar — experience and skills are yours to extend in mk-page.ts.

scholar.profile

The Bronx High School of Science / Neuroscience Dept, Columbia University / NIURA

Patents & publications on Google Scholar

Patents & publications

sourced from Google Scholar · in-ear EEG / neurotech

  1. Charging and processing case for wireless earbuds with in-the-ear electroencephalography implementation

    R Ahmed, S Huda, A Das, S Rajapaksha, M Shrestha, A Karim, C Kan, …

    US patent application · US Patent App. 18/459,379 · 2023

  2. Earbud apparatus with integration of real time in-the-ear electroencephalography and electrode port that can simultaneously play audio via speaker housing

    RF Ahmed, S Huda, A Das, S Rajapaksha, M Shrestha, A Karim, C Kan, …

    US patent application · US Patent App. 18/452,526 · 2024

  3. Electrode system for rubber ear tips with conductivity from n-doped silicone or conductive filaments in mixture for electroencephalography

    RF Ahmed, S Huda, A Das, S Rajapaksha, M Shrestha, A Karim, C Kan, …

    US patent application · US Patent App. 18/454,063 · 2023

  4. In-ear electroencephalography electrodes with multi-parameter vitals monitor connectivity

    RF Ahmed, S Huda, A Das, S Rajapaksha, M Shrestha, A Karim, C Kan, …

    US patent application · US Patent App. 18/452,561 · 2024

Experience

  • Research & prototyping

    NIURA · in-ear EEG · —

    Co-inventor on wireless earbud EEG patent family (see Scholar). Add titles, dates, and impact metrics here.

Education

  • The Bronx High School of Science

    Per Google Scholar affiliation

  • Columbia University

    Neuroscience Dept (affiliation on Scholar)

Skills

Signals & neurotech
In-ear EEG · Electrode systems · Motion artifact awareness
Build & research
Prototype validation · Med-tech rigor · Technical writing
Stack
TypeScript · Next.js · Python (add level)

// fill in — edit src/data/mk-page.ts

Nerd sheet

Metadata, honestly

The stuff I'd put in a README about myself if READMEs were allowed to be weird.

~/nerd_sheet.json

stack
Next.js · GSAP · Lenis · TypeScript
domain
med-tech · signals · validation
scholar
Google Scholar · patents
// in-ear EEG portfolio
signal_interest
ECG · SNR · motion artifact rejection
editor
Cursor (Geist enjoyer)
content
Instagram · TikTok · long-form research memos
github
melanikshrestha-boop
reading
systems papers · founder postmortems · FDA-adjacent docs
currently
clinical signal infra + RBCivosocial research thread
hot_take
a wrong abstraction costs more than a delayed launch
easter_egg
click anywhere — a soft glow follows
// subtle

Operate

How I work

Rigor before scale

In med-tech, a wrong abstraction costs more than a delayed launch. I optimize for correctness first, then velocity.

Build to learn

Code is an instrument for testing hypotheses. I ship prototypes to understand problems, not to perform building.

Make the next person faster

Every tool, doc, and system should reduce friction for whoever comes after — including future me.

Spec

System sketch

Block diagram — the nerdy footer for this page.

status = "online"

mode = "luxe"

cursor_tracking = "enabled"

scroll_chapters = "6"

last_build = "local"

SignalProcessOutputvalidation layer