I used to do a fair bit of astrophotography in university - it’s harder to find good skies now living in the city. Here are some of my old pictures. I’ve kept making rookie mistakes (too much ISO, not much exposure time, using a slow lens, bad stacking, …), for that I apologize!


ISS Transit (it was super fast, but I managed to get this):

Milky Way

Solar System

Sun, sunspots, prominences, filaments, Mercury (taken during the solar transit), Venus at half phase, Moon (blue filter), Mars (opposition), Jupiter with Io, Europa, Ganymede, Callisto, Saturn with its moons, Uranus and moons, Neptune and Triton.

Jupiter with the Great Red Spot.

Orion Nebula

The Pillars of Creation

I’m so unreasonably proud of this even though it looks horrible. You can just make out the pillars on the left.

Fireworks Galaxy with a Supernova

SN2017EAW. I also published this in a supernova catalog.


One hour of exposure.

Ring nebula, M13, Crab nebula

Star Trails

My first venture into astrophotography (Zambia). The picture on the right was taken during a meteor shower (can’t remember which one, but one is visible on the top right).

Einstein Cross

One of the leads of CUAS helped me locate the cross (thanks James!). I love this picture.

Comet Neowise

From my light-polluted London flat.



A lot of these (especially deep sky stuff) were taken using the 12” Northumberland telescope at Cambridge (I must’ve filled at least 25-33% of the logs the year I was there!). The Northumberland is a slow scope. I’ve also got a portable 4” scope that I’ve used for Andromeda and some planetary photography. Saturn was taken using a 16” scope at Harvard.


Nikon D3300, although I also use the advanced camera feature on my phone a lot.


Stuff that I actually use: a solar filter that I can’t find anymore, 25mm, 9mm eyepieces and phone/camera/telescope mounts.

Stuff I wish I used more: a focal reducer that has helped take pictures of some sparrows (causes chromatic aberration or distortion), some planetary and one moon filters, a special filter for spectrum analysis (that I’ve used once facepalm, for QSO 3C 273).


Photoshop for touch-ups, RegiStax and Deep Sky Stacker for stacking (that I run with wine… I hate this process so much, lemme know of alternatives!)

A Random Photo

A friend took this photo using the observatory’s solar telescope. Looks like an album cover!


Efficient Gaussian Process Computation

I’ll try to give examples of efficient gaussian process computation here, like the vec trick (Kronecker product trick), efficient toeliptz and circulant matrix computations, RTS smoothing and Kalman filtering using state space representations, and so on.

4 min read

Gaussian Processes in MGCV

I lay out the canonical GP interpretation of MGCV’s GAM parameters here. Prof. Wood updated the package with stationary GP smooths after a request. I’ve run through the predict.gam source code in a debugger, and mainly, the computation of predictions follows:

~1 min read


I wanted to see how easy it was to do photogrammetry (create 3d models using photos) using PyTorch3D by Facebook AI Research.

1 min read

Dead Code & Syntax Trees

This post was motivated by some R code that I came across (over a thousand lines of it) with a bunch of if-statements that were never called. I wanted an automatic way to get a minimal reproducing example of a test from this file. While reading about how to do this, I came across Dead Code Elimination, which kills unused and unreachable code and variables as an example.

~1 min read
Back to Top ↑



I used to do a fair bit of astrophotography in university - it’s harder to find good skies now living in the city. Here are some of my old pictures. I’ve kept making rookie mistakes (too much ISO, not much exposure time, using a slow lens, bad stacking, …), for that I apologize!

1 min read

Probabilistic PCA

I’ve been reading about PPCA, and this post summarizes my understanding of it. I took a lot of this from Pattern Recognition and Machine Learning by Bishop.

1 min read

Modelling with Spotify Data

The main objective of this post was just to write about my typical workflow and views rather than come up with a great model. The structure of this data is also outside my immediate domain so I thought it’d be fun to write up a small diary on making a model with it.

6 min read

Morphing with GPs

The main aim here was to morph space inside a square but such that the transformation preserves some kind of ordering of the points. I wanted to use it to generate some random graphs on a flat surface and introduce spatial deformation to make the graphs more interesting.

2 min read

SEIR Models

I had a go at a few SEIR models, this is a rough diary of the process.

4 min read

Speech Synthesis

The initial aim here was to model speech samples as realizations of a Gaussian process with some appropriate covariance function, by conditioning on the spectrogram. I fit a spectral mixture kernel to segments of audio data and concatenated the segments to obtain the full waveform. Partway into writing efficient sampling code (generating waveforms using the Gaussian process state space representation), I realized that it’s actually quite easy to obtain waveforms if you’ve already got a spectrogram.

5 min read
Back to Top ↑


Gaussian Process Middle C

First of my experiments on audio modeling using Gaussian processes. Here, I construct a GP that, when sampled, plays middle c the way a grand piano would.

1 min read
Back to Top ↑


Back to Top ↑