Content automation

Autonomous AI Video Studio — A Production System from Research to Publishing

End-to-end AI video production from research through publish—topic locks, cadence rules, and channel-ready output without chaining disconnected SaaS tools.

Developed by: TankDev
Project type: Internal product / production system
Field: AI content automation · video production
Status: Active production system
Outputs: Short-form · long-form video
Runtime: A structure that can run 24/7 independently of a local computer

System scope

These are defined production and control boundaries, not performance KPIs.

08Production stages

04API budget controls

02Distinct production flows

01Central control point

Problem and context

Manually managing topic research, source review, scripting, voice, visuals, editing, captions, quality checks, metadata, and publishing preparation across services creates a fragmented operation that is difficult to scale.

What TankDev built

A web-based production system that connects research, sources, script, voice, visuals, edit, checks, and publishing preparation in one workflow.

Engineering approach

Long-form and short-form content use distinct production logic. Scene timing follows generated voice duration; visual direction follows script and scene context; production rules are checked before publishing.

Publishing and platform boundaries

Supported YouTube production and publishing operations can run automatically through its API. For Instagram Reels and TikTok, vertical video, captions, titles, and publishing materials can be prepared. Automatic publishing scope depends on platform APIs, account type, and the access permissions currently available.

Control and ownership

Topic, audience, cadence, model providers, format rules, and budgets can be configured. The system can run 24/7 on a VPS without a local computer remaining on; keys, channel connections, settings, and media can remain in the project’s dedicated environment.

Current boundaries

Automatic publishing depends on channel APIs; final Reels and TikTok upload is manual. Unmeasured outcomes are not presented as performance claims.

Screens from the working production system

These screens are from the working Autonomous AI Video Studio management interface. Provider configuration, budget guardrails, publishing schedules, and connection checks are managed in the same operational console.

Operational visibility and readiness checks

The management navigation brings production, providers, budgets, scheduling, and YouTube settings into one place. The overview shows spend and production cadence, while the readiness layer checks required services before a run begins.

Autonomous AI Video Studio navigation showing videos, API keys, providers, budgets, scheduling, and YouTube settings. — Central production and settings navigation

AI video automation dashboard cards showing 24-hour, seven-day, and monthly spend together with weekly long-form and Shorts activity. — Period spend and production-cadence overview

Preflight screen showing readiness and connection latency for OpenAI, Gemini, ElevenLabs, YouTube, music, and database tooling. — Provider and connection readiness before production

Model providers and budget guardrails

Text, research, image, voice, music, and YouTube layers can be configured independently. Per-video, daily, weekly, and monthly caps prevent uncontrolled API consumption.

Provider configuration screen for text, image, voice, music, YouTube, and task-specific AI models in Autonomous AI Video Studio. — Task-specific provider and model selection

Configuration panel for long-form, Shorts, daily, weekly, and monthly API budgets together with concurrency and retry limits. — Configurable per-video and period budget limits

Publishing cadence, YouTube, and captions

Long-form and Shorts production run on separate schedules. YouTube privacy, API-audit state, and multilingual caption options remain explicit in the same management layer.