
_
critical rationalist
self-studying mathematics, computation, and explanations
About
Hi! My name is Sahil from India. I am self-studying mathematics and computer science with a research-oriented mindset. My work is guided by a strong interest in David Deutsch's four strands of explanations, which inform how I approach learning and problem-solving and life in general.
I build learning tools as a means of understanding deep concepts. Each project is an opportunity to test ideas, refine explanations, and develop better mental models.
Skills
Philosophy
My approach to life, learning and research is grounded in critical rationalism. I believe that knowledge grows through a process of conjecture, explanation, testing, and refinement; not through appeals to authority or credential-seeking.
- Explanations over authority: understanding trumps memorization
- Mathematics and computation as tools for building explanatory models
- Learning is iterative: conjecture, test, refine, repeat
- Credentials document past work; explanations demonstrate current understanding
Work & Collaboration
Content Collaboration
- Working with a content creator on content ideation and conceptual development
- Vibe-coding projects: combining aesthetics with functional implementation
Ongoing Work Projects
- Developing a learning framework for homeschooling based on Deutsch and Popper's epistemology: conjecture-driven inquiry and explanation-centered pedagogy. Check out CETR to know more.
Volunteer Roles
- IBM Qiskit Advocate: supporting quantum computing education and community engagement
- Gauss Scholar: contributing to mathematical education initiatives
Projects
Private Math Library
Private / In ProgressA comprehensive mathematics library implemented in Haskell and formalized in Lean. Aimed at students learning mathematics through verified proofs and computational examples.
Haskell, Lean, Type Theory
Learning Haskell
ActiveExploratory exercises and implementations to understand functional programming paradigms, type systems, and category theory through practical application.
Haskell
Python Blockchain
CompleteEducational blockchain implementation from first principles to understand distributed consensus, cryptographic primitives, and peer-to-peer systems.
Python, Cryptography
ECDSA Node
CompleteExploration of elliptic curve cryptography, digital signatures, and their applications in secure communication protocols.
Node.js, Cryptography
Raspberry Pi Camera Project
CompleteHardware integration project combining computer vision, real-time processing, and embedded systems programming.
Python, OpenCV, Raspberry Pi
Web3 / DApp Experiments
OngoingDecentralized application experiments exploring smart contracts, blockchain interaction, and distributed systems architecture.
Solidity, Ethereum, Web3.js
Notes
A personal research notebook. Emphasis on explanation over formalism.
- Definition
💡Explanatory Power
An explanation has explanatory power when it accounts for observations not by fitting parameters to data, but by providing a hard-to-vary account of the underlying reality.
For instance, consider the equation . This is not merely a formula that fits experimental data; it reveals a deep relationship between energy , mass , and the speed of light .
The explanatory power lies in the fact that this relationship is constrained by the structure of spacetime itself, not by arbitrary parameters.
- Note
λSK Combinators and Computational Universality
Schönfinkel and Curry discovered that all computable functions can be expressed using just two primitive combinators: and .
The combinator distributes application, while discards its second argument. Together, they form a complete basis for computation; any λ-term can be translated into SK combinator terms.
This reveals something profound: computational universality does not require many primitives. Complex behavior emerges from the composition of extremely simple rules. Wolfram's computational equivalence principle generalizes this: simple programs can be as powerful as any computation.
- Conjecture
◈Wolfram Physics Project and Hypergraphs
The Wolfram Physics Project proposes that spacetime emerges from the evolution of hypergraphs: abstract networks where nodes can be connected by edges linking more than two nodes simultaneously.
Starting from simple rewrite rules applied to these hypergraphs, the project aims to derive fundamental physics: general relativity, quantum mechanics, and potentially quantum gravity. The computational irreducibility of these systems means we cannot predict their behavior analytically; we must simulate.
A key insight is causal invariance: if different sequences of rule applications lead to equivalent states, this corresponds to reference frame independence in physics. The observer's perception of space and time emerges from their computational perspective within the hypergraph.
This suggests our universe is fundamentally discrete and computational at the Planck scale, with continuous spacetime as an emergent approximation. If true, it would unify computation and physics: the universe does not compute. It is computation.
- Note
⇌Isomorphism and Structural Identity
Two structures and are isomorphic if there exists a bijection preserving all relations and operations:
This definition captures structural identity: isomorphic structures are indistinguishable from the perspective of their internal relations. But what does this mean for mathematical truth?
If we accept the invariance principle (that mathematical properties should be invariant under isomorphism) then statements distinguishing isomorphic structures are not truly mathematical. This motivates category theory: objects matter only up to isomorphism, and morphisms (structure-preserving maps) become primary.
Consider the natural numbers. We cannot ask “is 3 a set?” in a mathematically meaningful way, because any isomorphic copy of has equal claim to being “the” natural numbers. Structure, not substance, determines mathematical identity.