AI Engineer (End-to-End Model Training & Deployment)

Overview

We are looking for a highly skilled AI Engineer to lead end-to-end AI-driven animation model development and deployment. This role requires hands-on expertise across the full pipeline — from dataset preparation and model training to workflow integration and serverless deployment.

The ideal candidate is capable of independently managing AI model development for image, video, and 3D animation use cases, and integrating trained models into production-ready systems.

Key Responsibilities

1. End-to-End Model Development

• Maintain and improve proprietary Character LoRA models.

• Perform LoRA image training and fine-tuning across:

Image generation models

Video generation models

3D-related models (where applicable)

• Develop AI-generated video pipelines, including:

Speech synchronization (lip sync)

Story and scene-building techniques

Animation-specific generation workflows

• Stay up to date with the latest open-source base models (e.g., Qwen, Wan, and other emerging models) and evaluate their applicability.

2. Dataset Engineering

• Collect, curate, generate, and refine datasets from:

Internal animation video assets

3D models and assets

Image libraries

• Build structured datasets suitable for:

LoRA fine-tuning

Video model training

Multi-modal model adaptation

3. Model Training & Infrastructure

• Independently execute full training pipelines:

Data preprocessing

Model configuration

Training and validation

Checkpoint management

• Train models using on-demand GPU services (e.g., RunPod).

• Optimize training for cost, performance, and scalability.

4. ComfyUI Workflow Development

• Design and configure ComfyUI workflows for:

Applying trained models

Modular AI animation pipelines

• Integrate trained models into ComfyUI nodes and workflows.

• Ensure production-ready stability and reproducibility of workflows.

5. Deployment & Platform Integration

Integrate trained models and ComfyUI workflows into the existing AI platform.

Add model weights and dependencies into Docker configurations.

Deploy models to RunPod serverless infrastructure.

Implement serverless API calls to retrieve inference outputs.

Maintain scalable, production-grade inference pipelines.

Required Skills & Qualifications

Technical Expertise

• Strong experience in:

LoRA training and fine-tuning

Image/video generative models

Multi-modal AI systems

• Hands-on experience with ComfyUI workflow design.

• Experience with on-demand GPU services (e.g., RunPod).

• Knowledge of:

Docker containerization

Serverless deployment architectures

API-based inference systems

Development Skills

Understanding of AI workflow orchestration.

Experience with web-based applications (preferably Next.js).

Familiarity with backend serverless infrastructure (e.g., RunPod serverless).

Ability to integrate AI models into production systems.