TEMPORAL CONTACT

Inspiration

What if you could call someone from the year 7444? Not a chatbot. Not a scripted character. A fully generated AI entity — with a face, a name, a job, and a perspective shaped by thousands of years of human (and non-human) evolution.

That question became TEMPORAL CONTACT.

What it does

TEMPORAL CONTACT is a time communication simulator. Users select a future year (3000–9999), choose a location anywhere in the universe, and connect with a randomly generated entity from that era.

Every session generates a unique being:

• 20+ persona types (scientist, warrior, diplomat, oracle...)
• 3 categories: Biological Human, Hybrid, Non-Human
• Rarity system: Common, Rare, Ultra Rare
• AI-generated avatar face locked for the entire session
• Video calls using image-to-video AI
• 10 languages supported
• Typewriter effect, sci-fi sound system, ambient music

The entity speaks, reasons, and responds — fully in character from their era. They treat 2026 as ancient history. Users can send images; the entity analyzes them as ancient artifacts.

How we built it

Built entirely on MeDo.dev no-code AI platform using:

• Large Language Model plugin for conversational AI
• Image Generation plugin for AI avatar creation
• Image-to-Video (Kling) for realistic video calls
• Text-to-Speech for entity voice
• Web Audio API for sound effects and ambient music
• Custom CSS for cinematic dark sci-fi UI

No traditional coding. Pure AI-powered no-code development with carefully engineered prompts.

Challenges we ran into

• Maintaining entity consistency (same face) across video calls required a FACE LOCK system
• Getting LLM to respond in correct language every time
• Preventing hardcoded template responses — making each entity feel truly unique
• Browser autoplay restrictions for background music
• Balancing immersion with technical limitations of no-code platform

Accomplishments that we're proud of

• 1 million+ possible unique entity combinations
• True session consistency: same name, face, persona throughout entire conversation
• Cinematic UI that feels like a real classified government terminal
• First Discord user reaction within hours of launch: "Cool project! Checked it out 👍"
• Built and iterated entirely through natural language prompts — zero traditional code

What we learned

• No-code AI platforms can produce production-quality experiences when pushed creatively
• Entity consistency is the hardest UX problem in generative AI experiences
• Sound design transforms a good UI into an immersive world
• Users engage longer when they don't know what to expect next

What's next for TEMPORAL CONTACT

• Mobile app version
• Multiplayer: two users connecting through same entity
• Entity memory across sessions (entity remembers you)
• Community gallery of encountered entities
• Voice input — speak directly to the future

Built With

• css
• image-generation
• image-to-video
• kling-ai
• large-language-model
• medo.dev
• text-to-speech
• web-audio-api