Skip to content
View ToldByNun's full-sized avatar

Block or report ToldByNun

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.

Content in all repositories owned by your account will be closed.
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
ToldByNun/README.md

header

typing


divider

About

Full-stack engineer building AI-powered tools, desktop applications, and developer-facing products — end-to-end, from UI to deployment. Python, TypeScript, Node, and Electron are my daily drivers; LLM orchestration and local-model tooling are where I spend most of my time.

I also have an uncommon background in low-level systems — CUDA kernels, Windows drivers, reverse engineering. That work is personal R&D, not my career target: what it gives me is an edge. I know what the machine is doing under my abstractions, so my high-level code tends to be faster, safer, and much easier to debug when things get weird.

Looking for full-stack / AI engineering roles on ambitious product teams.


featured demos banner

Demos

Sentinel
Deep learning engine in C++ & CUDA enabling full-model training and fine-tuning of LLMs directly on consumer-grade hardware.


Direct Link
Windows UEFI Bootloader Hook
Early boot execution and kernel-level hooking in a controlled environment — personal systems R&D.


Direct Link

Technical Stack

Daily Drivers
Python   TypeScript   JavaScript   Node.js   Electron



Tooling & Infra



Systems Foundation (personal R&D)
C   C++   CUDA   Lua

Metrics

stats    langs

activity


divider

Featured Projects

Mango

AI-driven Agentic Coding Framework tailored for Small Language Models (SLMs) — maximizing reasoning, tool execution, and context planning on lightweight, local models.


Job2CV

Cross-platform Electron desktop application for dynamic candidate-job matching, profile scoring, and local AI-powered resume customization.


Sentinel

Deep learning engine written in C++ & CUDA enabling full-model training and fine-tuning of Large Language Models directly on consumer-grade GPU hardware.


Sentic AI Platform

Production backend powering sentic.ai — Bayesian reasoning engine with custom search and multi-provider LLM orchestration.



Deep Systems & Security Research

Personal R&D I do to stay close to the metal — this is where my systems intuition comes from. Hobby, not job profile.

Project What it is
kernel-read-driver Windows kernel-mode driver for low-level memory operations and PEPROCESS CR3 retrieval.
UEFI Bootloader & Kernel Bridge Pre-boot execution module exposing controlled Ring -1 memory access interfaces.
Windows UEFI Bootloader Hook Proof-of-concept for early-stage system hooking in a controlled environment.
RE-UE4SS_lv-extension Live reverse-engineering engine for Unreal Engine apps with integrated disassembler.
Spectre-Memory-Framework Zero-dependency C memory library using ASM kernel hooks and ntdll redirection.
Chicony Spatial Tracking Low-latency positional tracking & prediction algorithm in C++, reconstructed from academic papers.

Engineering History

Timeline of shipped projects — view the full portfolio at toldbynun.github.io.

Date Project Description
Aug 16, 2026 Mango Agentic AI coding framework optimized specifically for Small Language Models (SLMs).
Jul 24, 2026 Sentinel High-performance C++/CUDA deep learning engine for local LLM fine-tuning and training.
Apr 27, 2026 Job2CV Electron platform for automated job discovery, match scoring, and AI document tailoring.
Feb 21, 2026 Sentic AI Platform Bayesian reasoning engine with custom search and multi-provider LLM orchestration.
Jul 10, 2026 RE-UE4SS_lv-extension Live reverse-engineering engine for Unreal Engine apps with integrated disassembler.
Jun 12, 2026 UEFI Bootloader & Kernel Bridge Pre-boot UEFI architecture providing controlled Ring -1 memory access interfaces.
Apr 15, 2026 IL osu! Autonomous Agent Real-time agent architecture testing Imitation Learning against low-latency inputs.
Mar 25, 2026 Rhythia Engine Integration Real-time GLSL/HLSL shader pipeline integrated into Dear ImGui interfaces.
Jan 16, 2026 Modular UI Library Reusable, high-efficiency UI component library built for modular game development.

Open Source Repositories

Repository Language Details
mango-public Python Agentic coding framework engineered specifically for Small Language Models (SLMs).
sentinel C++ / CUDA Lightweight deep learning engine built for training and fine-tuning LLMs on consumer GPUs.
kernel-read-driver C Windows kernel-mode driver for low-level memory operations and PEPROCESS CR3 retrieval.
Chicony-AA-Algorithm C++ Reconstructed low-latency spatial tracking algorithm implementation in native C++.
MecchaChameleon-External-Cheat C++ High-performance memory inspection and external process manipulation tool.
osu-parser-cpp C++ Lightweight, zero-dependency file parser designed for high-throughput processing.
py-arch-x Python Static analysis CLI tool leveraging ast to detect unreferenced functions across codebases.

Proprietary & Research Projects

Sentic AI Platform
Python
Production backend repository powering sentic.ai's epistemic AI tooling.
Rhythia Graphics
C++
Graphics library for injecting custom HLSL/GLSL shaders directly into Dear ImGui pipelines.
Spectre-Memory-Framework
C
Zero-dependency C memory manipulation library utilizing ASM kernel hooks and ntdll redirection.

Open to full-stack engineering, AI tooling, and desktop application roles.

footer

Popular repositories Loading

  1. Parcae Parcae Public

    ᛈᚪᚱᚳᚪᛖ ᚹᚪᛏᚳᚻᛖᛋ — deterministic C++20 toolkit for Liber Primus

    C++ 10

  2. MecchaChameleon-External-Cheat MecchaChameleon-External-Cheat Public

    Open source external Meccha Chameleon cheat in c++

    C++ 7 1

  3. sentinel sentinel Public

    zero-dependency C++/CUDA DL framework focused on single consumer GPU's

    C++ 1

  4. mango-public mango-public Public

    public repo for the mango framework source

    Python 1

  5. Unified-AI-Provider-Orchestrator Unified-AI-Provider-Orchestrator Public

    A modular Python-based abstraction layer to orchestrate multiple LLM providers

    Python

  6. py-arch-x py-arch-x Public

    A small, readable command-line tool that scans a Python codebase with `ast` and reports functions that are likely unused.

    Python