phoenix
phoenixPhoenix - Your AI Assistant
Inactive session will be ended in 24 hours
End Session

AI Engineer / NLP Engineer for an AI BF/GF Product

  • Posted:
  • Proposals: 32
  • Remote
  • #4506220
  • Open for Proposals
Óscar P.Bisma I.Amaneet B.Samiullah S.Malik F. + 27 others have already sent a proposal.
  • 5

Description

Experience Level: Expert
Estimated project duration: Not sure
We are launching a product in the AI romantic companions / AI BF/GF niche and need a specialist to take over the AI development part.

The website, registration, design, menu and payment system are handled separately. This role focuses only on the AI core: dialogue models, AI characters, memory, user adaptation, realistic image generation and MVP AI architecture.

Required skills

LLMs for conversational AI products.
Testing and comparing open-source and commercial models.
Llama / Hugging Face / vLLM / Text Generation Inference.
System prompts for AI characters.
Prompt engineering, fine-tuning and LoRA.
Memory layer design: mem0, Qdrant, embeddings, RAG.
Python, Docker, PostgreSQL, Redis.
GPU infrastructure and model deployment.
AI image generation: FLUX.1 / Stable Diffusion / ComfyUI.
Consistent face setup for AI characters.
Moderation / safety filters for AI dialogues.

Tasks

Select and test LLMs for romantic AI dialogues, set up AI characters, design user memory, plan adaptation to the user’s communication style, and suggest a pipeline for realistic AI selfies.

Experience with AI companions, dating, chatbots, virtual characters, roleplay scenarios or conversational AI is a plus.

The candidate should be ready to complete a short test task.

Please include your stack, relevant experience, project examples, readiness to complete the test task and your rate / estimated project cost.

New Proposal

Create an account now and send a proposal now to get this project.

Sign up

Clarification Board Ask a Question

  • Syed Haider A.

    I would like to know what will define success for your MVP: the most natural romantic conversations, strong long term memory, fast response time, or highly consistent AI selfies? Prioritizing this early helps choose the right LLM, memory pipeline, and infrastructure instead of optimizing the wrong thing.

  • Muhammad Ather A.

    Hi Alisa,

    I have a few quick questions before preparing our proposal:

    1. Do you already have a preferred LLM provider (open-source, OpenAI, Anthropic, etc.)?

    2. Will the AI companion support text only in the initial release, or are voice and multimodal interactions planned from the start?

    3. Is the GPU infrastructure already provisioned?

    4. Could you share the expected timeline for the MVP and the approximate number of AI characters planned for the first release?

    Looking forward.