// portfolio.log
// ravjothbrar
Latest Habs Occasional Paper No. 100 published ↗ 
██████╗  █████╗ ██╗   ██╗     ██╗ ██████╗ ████████╗██╗  ██╗
██╔══██╗██╔══██╗██║   ██║     ██║██╔═══██╗╚══██╔══╝██║  ██║
██████╔╝███████║██║   ██║     ██║██║   ██║   ██║   ███████║
██╔══██╗██╔══██║╚██╗ ██╔╝██   ██║██║   ██║   ██║   ██╔══██║
██║  ██║██║  ██║ ╚████╔╝ ╚█████╔╝╚██████╔╝   ██║   ██║  ██║
╚═╝  ╚═╝╚═╝  ╚═╝  ╚═══╝   ╚════╝  ╚═════╝    ╚═╝   ╚═╝  ╚═╝

██████╗ ██████╗  █████╗ ██████╗
██╔══██╗██╔══██╗██╔══██╗██╔══██╗
██████╔╝██████╔╝███████║██████╔╝
██╔══██╗██╔══██╗██╔══██║██╔══██╗
██████╔╝██║  ██║██║  ██║██║  ██║
╚═════╝ ╚═╝  ╚═╝╚═╝  ╚═╝╚═╝  ╚═╝

Ravjoth Brar

> Student AI Researcher
Interpretable AI & LLM Systems
Ravjoth Brar
Scroll Down
|| ABOUT ME ||

01 > I'm Ravjoth Brar — a passionate student AI researcher at Haberdashers' Boys' School, current Head Boy (School Captain) and a recipient of the Aske Scholarship.

My work spans from research and ethics to real-world applications of AI.

 

02 > My key interest is making AI more interpretable, ethical, and accessible.

My work has spanned the technical and the philosophical:

   — Fine-tuning language models and building bespoke AI systems

   — Exploring how frameworks like Sikhism can guide AI regulation (pending peer-review)

I recently gave a TEDx talk to 400+ students on whether AI is closer to God than humanity, publishing the 100th occasional paper in my school's series on the topic.

 

03 > Willing to learn from everyone — I'm actively seeking internships, placements, and research opportunities where I can contribute technical depth, creative problem-solving, and a fresh perspective on AI's societal impact.

Let's collaborate on building the future of responsible AI.

Per aspera ad astra
~ Seneca
Through hardship to the stars
|| EDUCATION ||
> Haberdashers' Boys' School
Secondary & Sixth Form Sep 2020 – Present
  • Appointed Head Boy (School Captain), visibly leading the entire student body (1000+) in assemblies, initiatives & meetings with the senior leadership team — first Sikh School Captain in school history.
  • Achieved all 10 Grade 9's (A**) in GCSEs, including a 100% score on DT coursework.
  • Studying Maths, Further Maths, Physics, and Philosophy at A-Level.

Roles / Activities

  • School Captain (Head Boy)
  • Student Council Vice Chair · Head of Digital · Head of Communication
  • TEDxYouth Speaker — AI and Religion (500+ audience)
  • Lead Scholar for CS & AI · Founder of AI Club · AI Policy Committee
  • Head of Model United Nations Club
  • Head of Sikh Religious Committee
  • Selected as 1 of 6 scholars in cohort for Academic Excellence Program
|| ROLES ||
> AI Intern
Internship Jul 2026 – Aug 2026
GFO-X
  • Built a dashboard to monitor token usage across the company, providing observability into spending habits and suggesting potential savings. Worked with another intern to:
  • Integrate the project with the relevant API keys and permissions
  • Normalise all data from these endpoints and store it in a lightweight SQLite database backend
  • Design the frontend in Python with Streamlit, displaying the necessary graphs
  • Use cron jobs to query the APIs and update the database at set intervals
> Cofounder
Part-time Oct 2025 – Aug 2026
Beri-Labs
> Research Writer
Part-time May 2026
Prismatic Studios
  • Researching, developing, and writing content to shape raw founder ideas into sharp, impactful brands
> AI Engineer
Part-time February 2026
The Web Kitchen
  • Engineered a high-performance (<500ms) RAG-based portfolio search tool using local embedding models to return relevant portfolio projects to streamline client workflow
  • Deployed a browser-based LLM (Qwen3-0.6b) to provide natural language justifications for project-client alignment
  • Implemented a hybrid search architecture that integrates vector similarity with metadata filtering, utilising sector-based boosting to increase recommendation accuracy
  • Deployed and secured OpenClaw - an autonomous agentic framework - in a Virtual Private Server, integrating it directly into Slack for automated SEO audits
  • Integrated the agent into internal task systems and GitHub repositories to log errors and autonomously make code edits in its own branch, before developer review
  • Conducted four red-teaming exercises to test agent security against indirect prompt injections, achieving a 100% bypass-resistance rate
> Founding Board Advisor
Part-time 2 years+ · Ongoing
Action Breaks Silence

A non-profit organisation dedicated to eliminating violence against women and girls globally

  • Led the Communications module for an ongoing charity initiative across continents
> Summer AI Intern
July 2025
Blankstate
  • Authored technical whitepaper on Intention Blended Framework (IBF), analysing how AI interprets organizational communication patterns for performance insights
  • Developed enterprise sales strategies across multiple channels
  • Contributed to ENTERPRISE 100 application, documenting privacy-first AI system that analyses data without storing sensitive information centrally
> Assistant Tennis Coach
Part-time · 3 years Jan 2023 – Jan 2026
Totteridge Tennis Club
  • Volunteered weekly for 3 years as a tennis coach
|| PROJECTS ||
> Towards Neural Interpretability: Proposing Transparent AI Frameworks to Predict, Localise, and Evaluate Epileptic Seizures
School (Extended Research Project) June 2026
  • The brain is the most complex object in the known universe - it's fundamentally a biological supercomputer - unravelling the brain's complexity represents an elusive challenge for humanity
  • Researching machine learning frameworks to model the brain, paving the path for neural interpretability
  • Developing and proposing novel applications of AI to not only predict epileptic seizures, but enable us to achieve transparent insights into how epileptic seizures are caused and how they propagate through the brain
Click here
> TEDx Speaker — "Is AI closer to God than humanity?" — Published Paper
Talk March 2026
TEDx Habs Elstree Youth TEDx talk thumbnail
▶
  • Delivered a TEDx talk to 400+ students exploring whether AI's high-dimensional cognition brings it closer to divinity than human understanding — drawing on Kant, AlphaGo, and the hard problem of consciousness.
  • Formalised my argument as the 100th article of the Haberdashers' School Occasional Papers Series
Watch on YouTube Read the paper
> NeuroLocate - Locating the Origin of Epileptic Seizures with a Physics-Informed Neural Network
Personal Ongoing
  • 20 million people worldwide suffer from drug-resistant epilepsy. For severe cases, the only remedy is removing where the seizures tend to originate from (the Epileptogenic Zone) with surgery, hence (as an extension of a research project) designed a:
  • Physics-Informed Neural Network that predicts where the Epileptogenic Zone is, locally in your browser
  • Integrated the 4 Hodgkin-Huxley equations (that describe the properties of the axon membrane) into the model's loss function to abide by brain thermodynamics
  • Lets you interact with a 3D WebGL model of the brain (and predicted seizure origin) with gestures
Click here
> Master the Member States - A Model UN Study Tool
Personal January 2026
  • Engineered an interactive Model UN study quiz to master flag knowledge and the official names of UN Member States (not 'Bolivia', but 'Plurinational State of Bolivia')
  • Ideation -> Deployment in 2 days
Click here
> YC Paris Startup School
Event June 29, 2026
Y Combinator · Paris, France
  • Attended YC's inaugural Paris Startup School at STATION F, Europe's largest startup campus. Heard from founders and investors shaping the next wave of technology:
  • Tom Blomfield — YC General Partner (Monzo founder)
  • James Hawkins — Co-founder & CEO, PostHog
  • Paul Copplestone — Co-founder & CEO, Supabase
  • Alexandre Lebrun — Founder, AMI
> Ode 2 Socrates
Personal March 2026
  • Ode 2 Socrates is a browser-based reflection tool
  • Built to help you look inward, question your assumptions, and form arguments with AI asking Socratic questions to challenge your thinking and interrogate your underlying assumptions.
  • It maps your thinking as a spatial graph, revealing connections and blind spots you didn't know were there.
  • "A space to find out what you think, how you think, and who you are."
Click here
> Building RIMICO - A Collection of Artisan Tools
Personal March 2026
Click here
> Fine-tuned AI Model for Biology Students
Oxford AI in Education Conference April 2025
  • Answering Biology GCSE questions can be hard. Exam-specific terminology means keywords and clear explanations are necessary, and generic AI tools often go too far in depth when answering biology questions
  • Fine-tuned SmolLM2-135M model using HuggingFace Transformers and PyTorch
  • Synthetically generated custom biology dataset adhering to A-Level and GCSE specification with labels
  • Implemented training checkpoints and mixed-precision training
> AI-powered Web App for Teachers
Personal February 2025
  • Web app which teachers at my school used to create subject-specific content such as worksheets and practice questions
  • Utilised the Google AI Studio API to provide responses
  • Constructed 5 prompt engineering courses, guiding students on how to effectively prompt AI
> Möbius Strip Topology in Neural Networks
Personal January 2025
  • Designed and implemented a novel neural network architecture incorporating Möbius strip topology to enhance feature learning and clustering
> How will Bias in AI affect society?
Essay January 2023
  • 3,500 word essay explaining how bias in AI arises and deliberating its potential impact on society in the next few years, and in the coming decades
  • Awarded a Distinction in my internal school competition
Click here
|| BLOG ||

>> Thoughts on AI, ethics, and the future as we know it.

> Latest Article
11/08/2026 > Claude's attempt at the Riemann Hypothesis An unreleased research version of Claude made an unexpected leap on the Riemann Hypothesis — raising the known lower bound from 41.6% to 67.2%. What happened, and what does it mean?
>> View all articles
|| PUBLICATIONS ||
Why AI is Closer to God than are Humans
Occasional Paper July 2026
Haberdashers' School Occasional Papers Series · Paper No. 100
  • > Argues that AI's high-dimensional cognition more closely resembles Kant's concept of Divine Intelligence than human understanding does — drawing on Locke, Russell, and Ghrist to establish the hard limits of human perception.
  • > Uses AlphaGo's Move 37 as a central case study: a move generated entirely through self-supervised learning that no human could explain, suggesting AI has transcended the phenomenal world into something closer to Kant's Divine Intellect.
  • > Demonstrates that AI satisfies both of Kant's criteria for Divine Cognition — atemporal processing (holding past, present, and future as simultaneous vectors) and modal realism (holding all possibilities in its probability space).
  • > Published as Occasional Paper Number One Hundred in the Haberdashers' School series.
View the Paper
Towards Neural Interpretability: Proposing Transparent AI Frameworks to Predict, Locate, and Evaluate Epileptic Seizures
Preprint June 2026
Zenodo
  • > Authored a preprint proposing novel AI frameworks to predict, locate, and evaluate Epileptic Seizures.
  • > Discussed how Liquid Neural Networks, Physics-Informed Neural Networks, and Topological Data Analysis can be combined to form a cohesive approach to modelling, and understanding, the epileptic brain.
  • > Awarded Distinction in my School's Extended Research Project.
View on Zenodo
> _
|| CV ||
|| CONTACT ME ||

>> Feel free to reach out — I'm actively seeking internships, research placements, and junior AI engineer roles for Summer/Autumn 2026.

email: ravjoth.brar@gmail.com
linkedin: linkedin.com/in/ravjothbrar
location: Barnet, London · UK
> _