Glossary AI & Vibecoding
Fundamental reference guide for the new era of software engineering by Andriy Orlov: autonomous agents, loop engineering, Model Context Protocol, context scaling, VPS infrastructure, and burnout prevention.
A
ENAPI Keys for Beginners (The Password to Model Intelligence)
A secret digital access token (starting with sk-...) that allows third-party applications (Cursor, plugins, Telegram bots) to interact with artificial intelligence without a web browser. Basic cyber hygiene rules.
API Usage Tiers
A tiered account system for developers using AI providers (OpenAI, Anthropic) that defines rate limits (RPM — requests per minute, TPM — tokens per minute) and daily financial caps that automatically increase with successful bill payments and security verification.
Automated Backup Recovery Testing
The practice of weekly automated deployment and verification of backups on ephemeral servers or containers, based on the principle: 'A backup does not exist until it has been successfully restored.'
C
ENCaddy & Traefik Modern Ingress Proxies
A new generation of web servers and ingress proxies that fully automate the issuance of TLS/SSL certificates, routing via Docker labels, and support for HTTP/3 protocol.
cgroups v2 Resource Quotas & OOM Watchdogs
Linux kernel mechanisms (Control Groups v2) for enforcing strict hardware limits on RAM, CPU, disk, and process counts (PIDs) to protect the host from runaway agent cycles.
Coolify (Self-Hosted PaaS)
An open-source infrastructure management platform (Self-Hosted PaaS, an alternative to Vercel, Heroku, and Render) that automates application deployment from Git, SSL certificate generation, database management, and backups on your own VPS.
Cron Schedulers & Systemd Timers
System daemons (Linux cron, systemd timers) and distributed queues (BullMQ, Temporal) that ensure guaranteed execution of periodic engineering tasks, backups, data synchronization, and AI agents on schedule.
D
ENData Leakage Through Chatbots
The risk of compromising corporate secrets, access passwords, and confidential personal data through inadvertent transmission to public cloud chatbots (ChatGPT, Claude, Copilot). Learn how to protect accounts and disable model training on your queries.
Disaster Recovery
A comprehensive engineering methodology and set of automated tools for creating immutable backups (RPO/RTO) with a guaranteed and regularly tested recovery protocol for system functionality.
Docker for Agents and Bots (Container Sandboxing)
A methodology for isolating autonomous AI agents, code interpreters, and background services in lightweight Docker sandboxes using cgroups and namespaces to prevent damage to the host OS.
E
ENeBPF Security & Runtime Auditing
A technology for safely executing bytecode directly within the Linux kernel (eBPF / Tetragon / Falco) to intercept dangerous system calls (syscalls), network connections, and file modifications by AI agents.
Edge AI in Smart Devices
The concept of executing machine learning algorithms directly on end physical devices (Edge Devices: smartphones, surveillance cameras, medical sensors, vehicles) without transmitting data to remote cloud data centers. This ensures instantaneous response in milliseconds, complete autonomy, and privacy.
Embedded Databases (SQLite & Turso / libSQL)
Embedded (In-Process) relational database technology based on SQLite and the distributed fork libSQL (Turso), combining operation without a dedicated network server with sub-millisecond read speeds.
Event-Driven AI with Webhooks
An architectural approach to Event-Driven AI leveraging Webhooks to instantly awaken language models in response to external triggers: a new payment in Stripe, a comment on Instagram, an email, or a message in Telegram.
F
ENG
ENGPU Slicing & Multi-Instance GPU (MIG)
This technology enables the hardware and software partitioning of a powerful GPU (NVIDIA H100 / A100 / RTX 6000) into multiple fully isolated instances, optimizing inference hosting costs.
Groq LPU Processors (500 Tokens Per Second)
The Groq Language Processing Unit (LPU) is a new type of specialized computing processor that generates responses from large models at speeds of 300–500 words per second by eliminating slow external memory (HBM) and utilizing ultra-fast SRAM. This architecture accelerates text generation to unprecedented levels.
H
ENHeadless Browsers (Playwright & Puppeteer)
A technology for programmatically controlling full-fledged browsers (Chromium, Firefox, WebKit) in the background without a graphical window for rendering complex SPAs, automated testing, and web agents.
Hetzner Bare-Metal vs Cloud Hyperscalers
A financial and technical analysis of infrastructure costs: why renting dedicated hardware in European data centers (Hetzner, OVH) is 5–10 times cheaper than AWS/GCP virtual machines for AI workloads.
Hourly GPU Rental in the Cloud (RunPod & Vast.ai)
Decentralized and cloud-based GPU rental platforms (RunPod, Vast.ai, Lambda Labs) enable developers and enthusiasts to rent powerful GPUs (Nvidia RTX 4090, A100, H100) with per-second or hourly billing, eliminating the need for expensive physical hardware purchases.
I
ENL
ENLitestream & Continuous SQLite Cloud Streaming
A technology for continuous background replication of SQLite's Write-Ahead Log (WAL) to cloud S3-compatible storage with zero data loss (RPO < 1 sec).
LLM Gateways & Routing (LiteLLM & Portkey)
Centralized engineering proxies for managing a fleet of models: automatic fallback between providers (Anthropic/OpenAI/Groq), semantic response caching, and budget quotas.
M
ENMicroVMs (Firecracker & Cloud Hypervisor)
Ultra-fast isolated virtual machines based on Linux KVM (Firecracker, Cloud Hypervisor) that start in 5–50 milliseconds for secure execution of agent code.
Model Fallback Chains
An architectural pattern for High Availability in AI systems. If the primary model provider returns a timeout error, exceeds rate limits (Rate Limit 429), or fails (Error 500), the system seamlessly switches the request to a backup model.
N
ENO
ENP
ENProduction-Grade Docker Compose Patterns
Engineering standards for secure deployment of multi-container services on VPS without the complexity of Kubernetes: resource limits, isolated networks, health checks, and secret management.
Prompt Caching
A technology for optimizing inference in large language models (Anthropic Claude, OpenAI, Google Gemini). It stores precomputed attention vectors of the static part of the input prompt (large books, codebases, or system instructions) in server memory, reducing the cost of repeated queries by 50–90%.
R
ENRate Limiting (Request Frequency Limitation and API Protection)
A systemic mechanism for controlling the intensity of incoming and outgoing traffic (Token Bucket, Sliding Window) to protect the backend from resource exhaustion, brute force attacks, Layer 7 DDoS, and financial overdraft on AI endpoints.
Rate Limits and Error 429 (Too Many Requests)
Provider-imposed restrictions on the speed and volume of requests to models (RPM — requests per minute, TPM — tokens per minute). This entry explains the causes of Error 429 and strategies to circumvent it.
Reverse Proxy (Nginx, Caddy, Traefik)
An intermediary server architectural layer that accepts external internet traffic (ports 80/443), performs SSL/TLS termination, compression (Brotli/Gzip), static caching, and securely routes requests to internal applications.
S
ENSecret Hygiene & Git Safety
A comprehensive set of engineering practices, cryptographic vaults, and pre-commit scanners (Gitleaks, Doppler, Infisical) for the secure management of API keys, tokens, and passwords without the risk of leakage into the public domain.
Serverless LLM Cold Start Optimization
A comprehensive set of engineering methods to reduce the cold start time of serverless containers with language models from 30–60 seconds to under 1 second: pre-warmed pools, chunked weight loading, and memory snapshots.
SSH Keys
An asymmetric pair of cryptographic keys (public and private) used by the Secure Shell (SSH) protocol for authentication without transmitting secrets over an unsecured network.
Streaming Text via SSE (Typewriter Effect)
A technology for transmitting generated tokens to the browser in real-time using the Server-Sent Events (SSE) protocol. It creates a typewriter effect, eliminating the unpleasant wait for a complete response.
T
ENTailscale & WireGuard Mesh Networks
A technology for building fully private overlay mesh networks without exposing public SSH or database ports to the internet using the modern cryptographic protocol WireGuard.
Telegram Bot API for Autonomous Agents
The official HTTP interface for the Telegram platform, enabling the creation of autonomous chatbots, AI assistants, Telegram Mini Apps (TMA), and interactive notification channels.
Token Pricing Math
A methodology for calculating financial costs associated with using commercial AI APIs. It explains the cost differences between input (Prompt/Input) and output (Completion/Output) tokens, hidden context costs, and the formula for estimating the unit economics of a startup.
U
ENV
ENVPS Hardening
A systematic process of configuring and reducing the attack surface of the Linux operating system on a virtual server through privilege restrictions, cryptographic isolation, and network auditing.
VPS Hosting
A model for providing isolated computing resources via a hardware hypervisor (KVM), offering full root access to a Linux operating system for deploying autonomous systems.
W
ENWebAssembly Sandboxing (Wasm / WASI)
A technology for isolated execution of third-party code and agent plugins within lightweight WebAssembly runtimes (Wasmtime, Wasmer) with sub-millisecond startup and zero access to the OS.
Webhooks (Asynchronous Webhooks)
An architectural pattern for asynchronous inter-service communication (Event-driven Push), where the event provider sends an HTTP POST request with data to a registered consumer URL upon the occurrence of a system event.
Z
ENZero-Downtime Deployment
A methodology and engineering mechanisms for updating production services without interrupting user service, breaking existing TCP connections, or generating HTTP errors 502/503.
Zero-Trust Architecture for AI Servers
The security paradigm 'Never Trust, Always Verify' eliminates static passwords, requires short-lived access certificates, mutual mTLS encryption, and complete isolation of secrets.
Full Stack of Tools, Tier Lists, and AI Skills
Beyond the glossary, GOTBURNOUT provides frontier AI benchmarks, a catalog of over 100,000 AI Agent Skills, production-ready VPS hardening scripts, and an exclusive Pro developer community.