Skip to content
View Samuele95's full-sized avatar

Block or report Samuele95

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Samuele95/README.md

Typing SVG

EmailWebsiteGitHubProfile Views


🧠 About Me

I'm an interdisciplinary researcher exploring how intelligent systems reason — from the symbolic structures of compilers to the emergent cognition of LLMs. My work seeks convergence between Neurosymbolic AI, Programming Language Theory, and Binary Analysis, drawing from mathematics, cognitive science, and social theory.

classSamuele95:
def__init__(self):
self.role="Interdisciplinary AI Researcher"self.education="MSc AI & Robotics @ UniCam"self.core_question="How do intelligent systems reason?"self.research= {
"neurosymbolic_ai": ["Reasoning Systems", "Context Engineering",
"LLM Cognition", "Symbolic-Neural Integration"],
"languages": ["Compilers", "Type Theory", "VMs", "Formal Semantics"],
"binary": ["Malware Analysis", "Reverse Engineering", "Binary Understanding"]
}
self.foundations= ["Mathematics", "Cognitive Science", "Neuroscience",
"Psychology", "Sociology", "Critical Theory"]
defphilosophy(self):
return"The deepest insights emerge at disciplinary boundaries"

Research Focus:Neurosymbolic AI · Reasoning & Cognition · Context Engineering · Compilers & VMs · Formal Language Theory · Binary Analysis

Deep Math Lover


🎯 Research Areas

Neurosymbolic AILanguages & CompilersMalware & Binary Analysis
Investigating reasoning mechanisms in LLMs through context engineering. Exploring symbolic-neural integration.Formal language theory, type systems, VMs, and the mathematical foundations of computation.Static/dynamic analysis, reverse engineering, program behavior at the lowest level.
How do machines think?How do we translate intent into execution?Where intent meets implementation.

Ψ NEOS — Neural Field Operating System

NEOS — The Operating System for Machine Intelligence

NEOS treats an LLM as a virtual machine and runs structured cognitive processes on it. No binaries. No file systems. Just meaning, dynamics, and reasoning.

Neural Fields · Symbolic Reasoning · Quantum Semantics · Observable Cognition

PaperWebsitePresentation


🔬 The Interdisciplinary Approach

I believe the most profound insights emerge at the boundaries between disciplines. My research integrates:

📐 Mathematics & Formal Theory🧬 Cognitive Science & Neuroscience
Type theory, category theory, formal verification, proof systems — the rigorous foundations of computation.How do humans reason, form concepts, and construct mental models? Biological intelligence illuminates artificial intelligence.
🧠 Psychology & Decision-Making⚖️ Law & Ethics
Cognitive and investigative psychology — rational choice theory, cognitive biases, uncertainty processing.Legal reasoning informs rule-based systems, precedent, interpretation, and AI governance.

📚 Sociology & Critical Theory — Criminal sociology and social theory through:

  • Foucault — Power structures & knowledge systems
  • Bentham & the Panopticon — Surveillance & control
  • Weber — Rationalization & bureaucratic reasoning
  • Le Bon — Crowd psychology & collective behavior
  • Malatesta — Anarchist theory & decentralized organization

🌐 The Convergence

 NEUROSYMBOLIC AI
Reasoning Systems
Context Engineering
|
+----------------+----------------+
| | |
v v v
MATHEMATICS COGNITIVE SOCIAL
Type Theory SCIENCE THEORY
Formal Systems Neuroscience Psychology
| | |
+----------------+----------------+
|
+----------------+----------------+
| | |
v v v
COMPILERS BINARY AI COGNITION
& LANGUAGES ANALYSIS LLM Reasoning
Formal Semantics Reverse Eng. Context Windows
  • Neurosymbolic AI bridges formal symbolic systems with neural learning
  • Context Engineering mirrors how humans construct meaning through contextual frames
  • Binary Analysis requires understanding both formal semantics and adversarial psychology
  • Compiler Design embodies the transformation of human cognitive structures into machine execution

🔬 Compilers, Interpreters & VMs

Compilation Pipeline:

  • Lexical Analysis — Tokenization, finite automata (Flex)
  • Parsing — LL/LR parsers, AST construction (Bison, ANTLR)
  • Semantic Analysis — Type checking, symbol tables
  • Code Generation — IR design, optimization passes
  • Runtime Systems — Memory management, GC, JIT

Virtual Machine Architecture:

  • Stack-based and register-based VM design
  • Bytecode instruction sets and encoding
  • Built LC3VM — Complete LC3 virtual machine

Language Design:

  • Formal grammar specification (BNF/EBNF)
  • Type system design and implementation
  • Built Logo4J — Logo language interpreter

CFlexBisonLLVMHaskell


🔍 Malware & Binary Analysis

Static AnalysisDynamic Analysis & Forensics
PE/ELF Analysis — Headers, sections, importsSandboxing — Isolated execution environments
Disassembly — x86/x64 with Ghidra & IDA ProAPI Monitoring — System call tracking
Pattern Recognition — Packers, crypters, signaturesMemory Forensics — Volatility, artifact extraction
YARA Rules — Detection signature authoringInjection Detection — Hollowing, hooking analysis
AI-Assisted — LLMs for code understandingProtocol RE — Understanding proprietary protocols

GhidraIDA Prox64dbgVolatilityYARA


⚡ Tech Stack

🤖 AI & Neurosymbolic Research

Claude CodeLangChainLangGraphClaude APIHuggingFace

🧠 Machine Learning & Deep Learning

PythonPyTorchTransformersNumPy

⚙️ Systems Programming

CC++JavaHaskellAssemblyLinux


🚀 Featured Projects

WebCatLC3VMLogo4JParkinson Repast Kit

📊 GitHub Analytics

GitHub StatsTop Languages

GitHub Streak


💬 Let's Connect

Interested in neurosymbolic AI, reasoning systems, compiler theory, or the intersection of computation and cognition?

Let's TalkPortfolio

Pinned Loading

  1. LC3VMLC3VMPublic

    Implementation of a LC3 Virtual Machine, with a simple Operating System, based on https://www.jmeiners.com/lc3-vm/

    C 16

  2. WebCatWebCatPublic

    Automated discovery and classification of websites content through unsupervised learning approach

    Python 3

  3. computational-fieldscomputational-fieldsPublic

    Interactive simulator for aggregate computing and self-organising programs — field calculus, building blocks, and real-time visualisation

    Python 1

  4. prometheusprometheusPublic

    Build-time meta-prompting framework for LLM agents. Treats a prompt as an operator, not a key — designs, audits, and scores it across 7 shapes and 3 modes, and ships a verifier.

    Python 1