Senior Python Backend Engineer (AI / Agents)

📋 Description

• • Vidrush Inc. is revolutionizing video production by developing an AI-native platform that leverages coordinated AI agents to replace traditional video teams. This innovative approach allows creators to transform a simple idea into a fully produced, long-form video—encompassing research, scriptwriting, voiceover generation, visual selection, motion graphics, and rendering—in under an hour. The platform aims to eliminate production timelines and editing complexities, enabling a seamless transition from concept to finished video.
• • As a Senior Python Backend Engineer (AI Agents), you will play a pivotal role in building and scaling Vidrush’s sophisticated AI-driven media pipeline. Your primary responsibility will be to develop and enhance an agentic pipeline capable of transforming user prompts into comprehensive video production workflows. This includes managing critical stages such as script generation, voiceover creation, footage selection, and video assembly. You will be instrumental in scaling the backend team to accelerate the execution of these ambitious goals.
• • Your day-to-day responsibilities will involve architecting and implementing agentic systems designed for reliable reasoning, planning, and the execution of multi-step tasks. You will be deeply involved in designing and deploying Large Language Model (LLM)-based services, focusing on functionalities like tool use, structured output generation, evaluation mechanisms, and robust monitoring. A significant part of your work will include developing advanced video and image intelligence pipelines, covering areas such as retrieval, matching, classification, metadata extraction, and generative augmentation.
• • You will also be responsible for integrating existing APIs and tools, demonstrating strong judgment in deciding when to build new solutions versus leveraging existing ones. A key aspect of the role is shipping scalable inference pipelines on cloud infrastructure, ensuring high performance and availability. You will apply best practices in MLOps and LLMOps, including implementing comprehensive observability, rigorous testing strategies, effective evaluation frameworks, version control, and diligent cost control measures. Furthermore, you will collaborate closely with the founders and product team to rapidly iterate on the platform based on valuable user feedback, ensuring the product stays at the forefront of innovation.
• • The team at Vidrush is characterized by its small size, high levels of trust, and a culture of significant ownership. There is a strong bias towards action and a relentless focus on shipping products efficiently. The team operates with lightweight specifications and embraces fast iteration cycles. A paramount focus is placed on developer experience, ensuring that testing is implemented where it truly matters, and observability is deeply integrated into all systems. This environment fosters rapid learning and professional growth, allowing engineers to make a substantial impact.
• • In this role, you will have the unique opportunity to contribute to a category-defining product at the intersection of AI, creator tools, and media production. You will gain hands-on experience with cutting-edge AI technologies, particularly in the realm of agentic systems and LLM applications for complex workflows. You will develop a deep understanding of building and scaling production-ready ML/LLM services, mastering MLOps/LLMOps principles. The position offers the chance to work with a passionate and driven team, contributing to a product that is fundamentally changing how video content is created. You will hone your skills in taking ambiguous problems and transforming them into tangible, shipped solutions, fostering a high degree of autonomy and impact.
• • This role is fundamentally about shipping and ownership. We are not seeking an engineer who prioritizes theoretical perfection or academic knowledge above all else. Instead, we are looking for an individual who can tackle ambiguous problems, move with speed, and consistently deliver tangible progress without the need for constant supervision. Your ability to translate complex challenges into working, scalable solutions will be paramount. You will be empowered to make significant technical decisions and drive projects from conception to deployment, contributing directly to the company's success and the evolution of AI-powered video production.
• • The ideal candidate will possess strong Python engineering skills, coupled with a proven track record of successfully shipping production systems. Experience in building with LLMs, including familiarity with OpenAI-style APIs, tool calling, Retrieval-Augmented Generation (RAG), and structured generation techniques, is essential. Furthermore, practical experience in deploying ML/LLM services in production environments, utilizing cloud platforms, containers, serverless architectures, and asynchronous pipelines, is highly valued. A solid, practical understanding of video/image processing, whether through Computer Vision (CV) or media pipelines, is also a key requirement. Above all, we are looking for individuals with high autonomy, capable of taking vague problems and transforming them into fully realized, shipped solutions, underpinned by strong engineering fundamentals such as clean code, reliability, performance optimization, and cost awareness.