Arush Gupta — Linux-Native Builder Who Self-Hosts Everything

+ 00:01:42 // LAT 20.5937° N

[ 35mm ][ 50mm ]

SYSTEM ACTIVE//DAILY-DRIVE ARCH LINUX

LOC:Based in India[2026]

ARUSH

GUPTA.

Linux-native builder who self-hosts everything.

Architecting sovereign bare-metal homelabs, orchestrating Podman containers, and running local quantized LLMs on llama.cpp. Zero third-party cloud reliance.

GITHUB PROFILE LINKEDIN [ View Systems ↓ ]

arush@archlinux:~

bash 5.2

arush@arch:~$neofetch --stdout

OSArch Linux x86_64

Kernel6.x-zen (Linux-native)

HomelabProxmox VE + Podman

AI Enginellama.cpp on-premise

Containers: RootlessUptime: 99.9%

// 01 ARCH-ROOTLESS // HARDWARE ACCELERATED

SCROLL TO DISCOVER SYSTEMS

[ STATUS: READY ]

[ 02 ]

// Philosophy & Background

ARCH // ZEN // SOVEREIGN

I am a Linux-native builder who daily-drives Arch Linux and self-hosts my entire digital world.

From bare-metal Proxmox hypervisors and rootless Podman containers to private llama.cpp inference engines, I reject proprietary black boxes in favor of open-source sovereignty, complete data ownership, and transparent system architecture.

Beyond the terminal, I channel the same obsessive dedication into cooking, fine-tuning audiophile sound equipment, gaming, football, swimming, and video editing.

DAILY DRIVER[PRIMARY]

Arch Linux

Minimal rolling-release installation running custom dotfiles, custom bash automation, and zen kernel for maximum responsiveness.

INFRASTRUCTURE[100% SELF-HOSTED]

Bare-Metal Homelab

Proxmox hypervisor hosting Podman rootless containers: Jellyfin, *arr stack, Immich, Vaultwarden, and llama.cpp local LLM inference.

OFF-SCREEN PURSUITS

Beyond the Terminal

// Creative passions & physical endurance

[HOBBY]

Cooking

Culinary craft & flavor experiments

[HOBBY]

Audiophile

Hi-Fi acoustic fidelity & gear

[HOBBY]

Gaming

Strategy & competitive titles

[HOBBY]

Football

Pitch tactics & weekend matches

[HOBBY]

Swimming

Endurance, laps & mental reset

[HOBBY]

Video Editing

Pacing, color grading & visual rhythm

ARCH LINUX

PROXMOX VE

PODMAN CONTAINERS

LLAMA.CPP

JELLYFIN

IMMICH

VAULTWARDEN

*ARR STACK

PYTHON

BASH SCRIPTING

JAVA

OPENCLAW

C++

LOCAL AI

ARCH LINUX

PROXMOX VE

PODMAN CONTAINERS

LLAMA.CPP

JELLYFIN

IMMICH

VAULTWARDEN

*ARR STACK

PYTHON

BASH SCRIPTING

JAVA

OPENCLAW

C++

LOCAL AI

ARCH LINUX

PROXMOX VE

PODMAN CONTAINERS

LLAMA.CPP

JELLYFIN

IMMICH

VAULTWARDEN

*ARR STACK

PYTHON

BASH SCRIPTING

JAVA

OPENCLAW

C++

LOCAL AI

[ 03 ]

// Technical Arsenal & Infrastructure

4 ARCHITECTURAL CATEGORIES

Built from scratch. Hosted on-prem.

Every layer of my stack—from low-level bash automation to containerized microservices and hardware-accelerated LLMs—is tailored for sovereignty, efficiency, and zero cloud dependency.

[01]

CORE LOGIC

Languages

Foundational programming and scripting languages powering tools, automation, and web interfaces.

Python

Automation, Telegram bot backend & toolingActive

Bash Scripting

System administration, dotfiles & container provisioningActive

Java

Object-oriented software development & algorithmsProficient

C++

Low-level system mechanics & foundational logicBasic

HTML / CSS

Modern layout architectures & responsive designLearning Full Stack

// VERIFIED & OPERATIONALREADY →

[02]

ON-PREM LAB

Self-Hosting & Infra

Complete sovereign bare-metal homelab infrastructure running isolated rootless containers and private services.

Proxmox VE

Type-1 bare-metal VM & LXC container orchestrationHypervisor

Podman Containers

Daemonless, secure OCI container architectureRootless

Jellyfin

Personal zero-telemetry media streaming serverDeployed

The *arr Stack

Automated media indexing & acquisition pipelinesAutomated

Immich

High-performance private photo & video backupProduction

Vaultwarden

Self-hosted zero-knowledge password vaultEncrypted

// VERIFIED & OPERATIONALREADY →

[03]

LOCAL INFERENCE

AI & Tooling

Running local language models on bare metal and architecting autonomous assistant backends.

llama.cpp

The foundational C/C++ backend that Ollama is built onCore Engine

OpenClaw Backend

Custom AI engine powering nightly Telegram chatbotIntegrated

Telegram Bot API

Nightly automated conversational check-insAutomated

Git & Version Control

Branching, repository maintenance & workflowsDaily

Linux CLI Utilities

tmux, SSH, systemd, rsync, curl, cronNative

// VERIFIED & OPERATIONALREADY →

[04]

DAILY DRIVER

Operating System

Deep Linux-native workflow configured for raw speed, minimal overhead, and complete operational transparency.

Arch Linux

Primary operating system with rolling updates & zen kernelDaily Driver

Linux Ecosystem

Process isolation, systemd services & storage primitivesNative

Bash Environment

Custom shell scripts, aliases & terminal keybindingsConfigured

Rootless Security

Unprivileged user namespaces & container hardeningEnforced

// VERIFIED & OPERATIONALREADY →

[ 04 ]

// Featured Builds & Infrastructure

3 ACTIVE PRODUCTION BUILDS

Systems running in production.

Real deployments running on real hardware. Zero mockups, zero simulated metrics—just hardened self-hosted software and custom tooling built for daily use.

01//INFRASTRUCTURE & ORCHESTRATION

ONLINE

Self-Hosted Homelab Infrastructure

A resilient bare-metal home server engineered on Proxmox VE. Employs rootless Podman containers to orchestrate a sovereign suite of privacy-first services: Jellyfin for media streaming, the *arr stack for automated acquisition, Immich for local photo/video vaulting, and Vaultwarden for zero-knowledge credential security.

Proxmox VEPodman ContainersJellyfinThe *arr StackImmichVaultwardenLinuxBash

VIEW ON GITHUB

SYS_LOG // HOMELABPODMAN RUN

$podman ps --format 'table {{.Names}} {{.Status}}'

jellyfin Up (healthy) 42 days

immich-server Up (healthy) 42 days

vaultwarden Up (healthy) 42 days

arr-pipeline Up (healthy) 42 days

[ 50mm // EXP 01 ]

02//ARTIFICIAL INTELLIGENCE & INFERENCE

ONLINE

Local LLM Hosting with llama.cpp

Hardware-accelerated local large language model hosting running on-premise using llama.cpp—the foundational C/C++ backend that Ollama is built on. Configured for quantized GGUF execution, low-latency token generation, and local API serving with 100% private prompt containment and zero third-party cloud telemetry.

llama.cppC++PythonGGUF QuantizationLocal LLM InferenceZero-Cloud Privacy

VIEW ON GITHUB

SYS_LOG // LLAMA-CPPPODMAN RUN

$./llama-server -m models/7B-Q4_K_M.gguf --ctx-size 8192

llama_init_from_file: model loaded successfully

offloading layers to accelerator: [ACTIVE]

HTTP server listening on 127.0.0.1:8080 (0.0ms cloud lag)

[ 50mm // EXP 02 ]

03//AUTONOMOUS AGENTS & AUTOMATION

ONLINE

Nightly Telegram AI Assistant

An autonomous conversational AI chatbot built with an OpenClaw backend and integrated directly into Telegram via the Bot API. Engineered to proactively message every single night with structured day recaps, scheduled check-ins, and contextual prompts without relying on closed hosted platforms.

Telegram Bot APIOpenClaw BackendPythonScheduled AutomationSystemd Service

VIEW ON GITHUB

SYS_LOG // TELEGRAM-BOTPODMAN RUN

$python3 -m openclaw.daemon --schedule '23:00'

[cron] cron trigger activated at 23:00:00 IST

[openclaw] generating structured daily recap...

[telegram] outgoing payload dispatched -> @arush [200 OK]

[ 50mm // EXP 03 ]