Physics Graduate · Aspiring Astrophysicist · Lifelong Learner
I'm someone who finds genuine wonder in the universe — from the mathematics behind a collapsing star to the quiet satisfaction of a well-written Python script. I've had the chance to work with NASA datasets, publish research on stellar flare prediction and geomagnetic storm forecasting, experience instruments at the National Physical Laboratory, and lead a team through the chaos and joy of a hackathon. This is a small corner of the internet where I share what I've been exploring.
I'm a physics graduate from Shivaji College, University of Delhi, with a deep love for astrophysics and the kind of questions that don't have easy answers. I'm drawn to the space where rigorous science meets human curiosity.
Over the past year I've had some really meaningful experiences — interning at Starscapes where I got to work on data visualization, spending time at the National Physical Laboratory in their Optical Radiation division, and leading a team at the NASA Space Apps Challenge to model 3D exoplanetary environments. Each of these shaped how I think about research.
I qualified IIT JAM 2026 in Physics, which felt like a meaningful step in a journey that's very much still unfolding. I'm excited about what comes next and always open to conversations about physics, astronomy, or anything in between.
Worked on astrophysical data analysis and visualization using Python and MATLAB. Also contributed to astronomy outreach content — which reminded me how much I enjoy making complex ideas accessible to people.
Spent time in the Optical Radiation and Metrology Division working on calibration techniques and precision measurements. It was my first real encounter with how careful and methodical scientific instrumentation has to be.
Led a team of four to model 3D exoplanetary environments using NASA open datasets. It was equal parts exhilarating and humbling — coordinating ideas, managing time pressure, and presenting something we built together from scratch.
Participated in workshops on cosmology, general relativity, and astronomical simulations. Being surrounded by researchers and fellow enthusiasts in that setting was genuinely inspiring.
Built an Arduino-based interactive setup using LEDs to demonstrate the speed of light and planetary distances. I wanted to create something tactile that could make abstract astronomical scales feel real — especially for younger students. It was one of the more satisfying things I've made.
A two-layer LSTM network trained on Solar Cycle 24 storm events achieves an overall MAE of 7.14 nT on the Solar Cycle 25 test set — a 79% improvement over persistence. SHAP attribution analysis identifies the failure mechanism for the May 2024 Gannon superstorm: extreme solar wind speeds absent from SC24 training caused systematic model misinterpretation. This work constitutes the first cross-solar-cycle generalisation evaluation for equatorial electrojet prediction at the Indian sector.
Real-time SYM-H prediction using live NOAA solar wind data. The model runs 24/7 on a home server in India, fetching DSCOVR satellite data every 5 minutes and predicting geomagnetic conditions 60 minutes ahead.
Whether it's a research opportunity, a question about astrophysics, or just a conversation — I'm genuinely happy to hear from you. Don't hesitate to reach out.
Send a Message