You can't keep me in here forever.
Jose
What's an Arun Jose?
Hello there. I'm Arun Jose.

I like making things, and I make whatever I think would be good, regardless of domain - I sometimes develop websites and apps, research things that seem interesting, write about whatever I want, and draw on the rare occasions I remember to.

I sometimes go by the mononym Jozdien. Don't ask me what that means, because I don't know either. ‚Äč

I want to try to solve the most important problems in the known universe - why aim small, right? - which I think is AI Alignment in this phase of human history.

I'm an aspiring rationalist, which means I try to think about how I think.

I enjoy all genres of fiction when done well, but my favourites are rational fiction and meta/absurd comedy (as well as crack fanfiction, which feels appropriate closed off in brackets).

Instagram and Email are my most frequented platforms for messages, so if you have ideas, want to collaborate, or just think I'd find something interesting (I probably will - err on the side of messaging me), just send me a message with the phrase "Xavier Avial".
Work
The Calibration Game
An android app that optimizes credence calibration - the accuracy of belief strengths against real success rates - with a scoring system that incentivizes alignment of "sureness" with success rates.

A better description can be found here.
React Native
Firebase
Heuristics
The Calibration Game
An android app that optimizes credence calibration - the accuracy of belief strengths against real success rates - with a scoring system that incentivizes alignment of "sureness" with success rates.

A better description can be found here.
React Native
Firebase
Heuristics
Metise
A conditional generative adversarial network for fixed image (de-)compression. Compression is performed along the dimensions of colour density, being a relatively intractable direction outside neural net reconstruction.

I'm currently writing a paper on this, but the code is available below.
TensorFlow
cGAN
Metise
A conditional generative adversarial network for fixed image (de-)compression. Compression is performed along the dimensions of colour density, being a relatively intractable direction outside neural net reconstruction.

I'm currently writing a paper on this, but the code is available below.
TensorFlow
cGAN
Mentem
Mentem was originally envisioned as an all in one platform for mental health support using virtual, professional, and community resources, triage using speech and facial analysis, and health clinic assistance. It is currently being developed into an all in one platform - a hospital ERP system.
OpenCV
SpaCy
WebRTC
Mentem
Mentem was originally envisioned as an all in one platform for mental health support using virtual, professional, and community resources, triage using speech and facial analysis, and health clinic assistance. It is currently being developed into an all in one platform - a hospital ERP system.
OpenCV
SpaCy
WebRTC
Karma
An emergency management framework that incentivizes small-scale demand-specific rescue operations. Emergency alerts are sent out on a tiered structure using P2P networks, ordered by severity. Damage extent maps are plotted using alert data.
PHP
Leaflet
Karma
An emergency management framework that incentivizes small-scale demand-specific rescue operations. Emergency alerts are sent out on a tiered structure using P2P networks, ordered by severity. Damage extent maps are plotted using alert data.
PHP
Leaflet
Veris
An open-source educational administration framework I helped build. This is being developed to replace outmoded methods of handling data and process in higher educational institutes, while providing the flexibility to fit specific use-cases.

Currently in alpha testing in CET.
NextJS
React Native
Veris
An open-source educational administration framework I helped build. This is being developed to replace outmoded methods of handling data and process in higher educational institutes, while providing the flexibility to fit specific use-cases.

Currently in alpha testing in CET.
NextJS
React Native
Phemos
A news algorithm I developed to account for a documented positive feedback relationship between media coverage and electoral polling. The algorithm weighs down overexposure on a log scale.

More on the aforementioned effect here.
Research
Python
Phemos
A news algorithm I developed to account for a documented positive feedback relationship between media coverage and electoral polling. The algorithm weighs down overexposure on a log scale.

More on the aforementioned effect here.
Research
Python
History
Positions
Co-ordinator/Co-Founder
  • Set up an industry collaboration with Neuroplex to provide experienced training and internship opportunities.
  • Hosted workshop series on Machine Learning, Deep Learning, and Friendly AI - 23 participants went on to complete projects in predictive analysis.
  • Hosted webinar series on Data Science and Deep Learning.
  • Organized AI Safety competiton, which closed with an analysis of the common fallacies faced (Check that link out, I swear it's cooler than it sounds).
  • Collaborated with the Machine Intelligence Research Institute for instructional material and technical assistance.
  • Founded 2 Special Interest Groups for AI in IEEE Student Branches across the state.
Intern - ML Team
Airli, OrbitAustralia · Since September 2020
Developing a machine learning model to process satellite imagery to generate actionable insights relating to pollen forecasts, for clients working to improve quality-of-life apropos of asthma and air quality.
Freelancer
Android & Web Development · Since August 2019
I do front-end design and development.
Logistics Manager
MuLearn · February 2019 - August 2019
  • Oversaw heuristics of instructional methods and handling of source material.
  • Managing and formatting the MuLearn website.
Awards
EVOKE'19 Hackathon
1st Place · February 2019
National level tech summit by IEDC TKMCE and IEEE SB TKMCE.
Developed an emergency management platform (Refer Karma in the preceding section).
Reboot Kerala Hackathon
3rd Place · March 2020
Healthcare Category, State level Hackathon series by the State Department for Higher Education.
Developed a mental health triage and support framework (Refer Mentem in the preceding section).
Pass the Code
1st Place · March 2019
State-level tag-team coding competition conducted by the SBs of IEEE, IEDC, and ISTE of Barton Hill College of Engineering.
AKCSSC Web Design Contest
1st Place; November 2020
State-level web design competition conducted by IEEE CSKS as part of AKCSSC 2020.
Submitted the NextJS-built web front-end of Veris (refer in the preceding section).
Oratrium
1st Place · October 2015
State level public speaking competition by JCI, judged and awarded by Dr. Shashi Tharoor.
Education
College of Engineering Trivandrum
B.Tech CSE · Class of 2022
When I'm not cutting classes to do something else, I'm a third year Computer Science undergraduate.
Loyola School, Thiruvananthapuram
ISC · Batch of 2018
There's much more to say than could fit in this space to do justice.
Art
Stuff I Write
Posts
The Truth is Often Comfortable
Jan 2021 · 5 min read
Utopic Nightmares
Jan 2021 · 8 min read
The Loyolite
Jan 2020 · 5 min read
Fiction
Think I'd find something interesting?
I probably will, I'm easily amused. If in doubt, send the message anyway, and preface it with "Xavier Avial".