Senior Applied GenAI Video Engineer
At a glance
Senior AI Engineering role at Atria. Remote, China · full-time.
Pay not stated AI Engineering salary
remoteroles summary, based on the employer's posting.
What you'll do
- Build the video pipeline from prompt and asset handling through shot assembly, rendering, finishing and delivery
- Turn product aims into trackable results for video quality, ad performance proxies, response time and cost
- Connect external video-generation services, handling releases, usage limits, backup paths and production reliability
- Set up ways to assess video quality and whether a creative is ready to run as an ad
- Improve speed, capacity and spend with caching, batching, retries, parallel work and content-aware routing
What you bring
- You have built video-generation systems that run in production
- You can engineer in Python, with strong software design, testing, review and performance-tuning skills
- You understand LLMs and multimodal prompts, including structured responses, tool use and guardrails
- You can work with CI/CD, containers, experiment tracking, monitoring and cloud deployment, and know AWS, GCP or Azure
Who this fits
This suits a hands-on senior engineer who wants to ship and operate video-generation systems, rather than stop at prototypes. The role is remote in China and calls for clear English communication in technical writing and collaboration.
From the employer
About Us
Atria is an AI-powered ad creative workflow platform designed to help marketing teams generate more revenue by producing winning ads more efficiently. Trusted by top brands like Ryze Superfoods, Solawave, and Loop Earplugs, as well as forward-thinking agencies such as Power Digital, Inspire Digital, and Disruptive Digital, Atria focuses on enhancing performance across Meta and TikTok channels.
About the role
You'll own the end-to-end video generation pipeline that powers ad creative production — from vendor selection and prompt/media orchestration to scalable serving, quality evaluation, and reliability in production. This is a hands-on engineering role focused on shipping systems that move business metrics, not just prototypes.
What you'll do
• Build and own video generation pipelines: prompt orchestration, asset conditioning, shot assembly, rendering, post-processing, and delivery.
• Translate product goals (better-performing ad creatives, faster iteration) into measurable GenAI outcomes (quality, CTR lift proxies, latency, cost).
• Integrate external video GenAI vendors/APIs and manage versioning, fallbacks, quotas, and reliability.
• Create evaluation frameworks for video quality and ad readiness.
• Optimize for latency, throughput, and cost (caching, batching, retries, parallelism, content aware routing).
• Partner with product/design/engineering to ship improvements quickly and safely.
What we're looking for
• Experience building production video GenAI systems
• Strong engineering skills: Python, solid software design, testing, code reviews, performance tuning.
• Practical understanding of LLMs and multimodal prompting: prompt design, tool/function calling patterns, structured outputs, guardrails.
• Familiarity with DevOps: CI/CD, containers, experiment tracking, monitoring/alerting, cloud deployment.
• Comfort with at least one major cloud: AWS/GCP/Azure.
• Able to communicate clearly in English (technical documentation and collaboration).
Nice to have (Plus)
• Video/media stack experience: FFmpeg, encoding/transcoding, timeline composition, asset management, CDN delivery.
• Ads/marketing domain knowledge: creative testing workflows, performance measurement, attribution basics.
What We Offer
• Fully remote environment with flexible schedule.
• Opportunity to work with cutting-edge GenAI video technology in a fast-growing company.
• Collaborative, inclusive culture with a small team and high ownership.
• Competitive compensation and benefits package.
• Strong opportunities for professional growth and development.
If you're excited about building scalable, reliable video GenAI systems that directly impact business outcomes, we'd love to hear from you.
Kind reminder: No duplicate applications for the same role please. Thanks.
AI Engineering work from LinkedIn, X, Reddit and Upwork, in one list.
Each morning you get everything that fits, strongest first.
AI EngineeringRemote
Find me workBy the team behind remoteroles.