AI Engineer, Agentic Ad Creative (Multimodal)

NewsBreak • Mountain View, CA

13d • Remote

About The Position

NewsBreak is seeking an AI Engineer to own the end-to-end development of an agentic creative pipeline. This role involves building and optimizing systems that generate thousands of images and videos daily for various advertising platforms. The goal is to create an agent that continuously improves its creative output based on performance feedback (CPI and ROAS), making the creative process more efficient and cost-effective.

Requirements

• Deep understanding of diffusion and video models, including debugging, training LoRAs, optimizing inference, and building ComfyUI workflows.
• Familiarity with various generative models (Flux, SDXL, Veo, Sora, Runway, Kling, Hailuo) and their respective strengths, weaknesses, and cost-performance trade-offs.
• Demonstrable public artifacts such as a fal demo, Replicate model, Civitai upload, viral X thread, or GitHub ComfyUI workflow.
• Strong instinct for creating scroll-stopping ad content that captures attention within the first 3 seconds.
• Experience considering cost as a primary design constraint (e.g., $/asset, $/click, $/install) and shipping cost arbitrage between model tiers.
• Thorough knowledge of Meta Advertising Standards and TikTok branded-content rules, treating policy as a pre-deployment gate.
• Proven ability to build working pipelines from idea to shipped product, iterating based on performance data.
• Experience debugging CUDA OOM errors or similar low-level issues related to generative models.

Nice To Haves

• Pretrained or significantly advanced a VLM.
• Open-sourced a ComfyUI custom node, diffusers extension, or Replicate model with significant traction.
• Deployed serverless inference on fal/Replicate/Modal at scale (>10k req/day).
• Shipped a video product end-to-end: script → scene plan → b-roll → voiceover → captions → final cut.
• Authored a public take on generative video model failure modes that is cited by other engineers.

Responsibilities

• Extend LLM critic and performance-feedback regeneration from images-only to images, video, and copy, operating unsupervised at thousands of variants per day.
• Build a video pipeline that generates scripts, scene plans, b-roll, voiceovers, and captions for platform-ready 9:16 vertical ads in under 90 seconds and under $0.50 per iteration.
• Develop a generative model router to select the optimal model (e.g., Flux Schnell vs Pro, Veo 3, Sora, Runway Gen-3/4, Kling, Hailuo) for each brief, minimizing cost per asset while maintaining a quality floor, and prove its effectiveness via A/B testing.
• Create a VLM (Vision-Language Model) to pre-flight check every frame and caption against current Meta, Google, and TikTok ad policies before upload, preventing policy violations at the source.
• Train and deploy brand-style LoRA/DreamBooth checkpoints on past CPI-winning ads, using serverless deployment on platforms like fal/Replicate, and A/B test against base models.
• Develop an event-triggered ad pipeline for video, enabling rapid creation of platform-specific ad variants with locale-specific voiceovers in response to real-time events.

Benefits

• Competitive Bay Area compensation and meaningful equity.
• Unlimited generative model budget (fal, Veo, Sora, Runway, ElevenLabs, OpenAI image, Anthropic vision, Hedra, Kling).
• Health, dental, and vision care for you and your family (100% coverage for employee).
• Top-tier 401(K) plan with company matching.
• Paid time off and paid holidays.
• FSA, HSA, and commuter benefits programs.
• Team activity budget.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed