Backend Engineer, AI (Agent Systems)

About the Company and Product Mission

Our mission is to construct an intelligent, proactive assistant tailored for daily users who currently engage with basic applications like email, notes, and task managers that lack AI integration. The aim is to enrich users' interactions by infusing intelligence into conversations, errands, organization, and workflows with minimal effort from the user side.

Product Focus

The product prioritizes achieving exceptional reliability in managing long-term workflows, maintaining persistent contextual information, and successfully completing real-world tasks. It requires handling multifaceted reasoning processes, communicating with external tools, and ensuring dependability despite the inherent unpredictability of AI models. Ultimately, the goal is to help users perform tasks more enjoyably and reduce completion time by about 90%.

Role Overview

As a Backend Engineer with a focus on AI, you will be responsible for the inference and orchestration system that underpins every AI-driven interaction within the product. This critical layer interfaces between the AI models and the end users, where factors such as response time, accuracy, reliability, and operational cost significantly affect the user experience.

You will build and maintain production-grade systems that convert AI model capabilities into fast, robust, and observable APIs accessible to both mobile and desktop applications.

Key Responsibilities

• Develop and manage backend infrastructure supporting AI functionalities in a production environment.
• Architect and implement inference pipelines, orchestration mechanisms, and define service boundaries connecting with models.
• Take full ownership of operational aspects including monitoring, logging, alert setup, and rapid incident response.
• Enhance system performance by optimizing latency and throughput through efficient inference methods, caching, batching, and streaming strategies.

Desired Experience and Skills

• Solid foundation in backend engineering with experience in production systems.
• Proven ability to operate services that require high throughput and low latency.
• Understanding of AI inference methodologies including large language models (LLMs), embeddings, and multimodal inputs.
• Competence in diagnosing and troubleshooting distributed system issues, especially under load.
• An outcome-driven mindset that emphasizes shipping features quickly and learning from their production use.

Expected Outcomes

• Reliable backend systems that scale to meet production AI traffic demands with minimal latency and maximal throughput.
• Stable and clear APIs designed for seamless integration with frontend clients and machine learning systems.
• Swift detection and resolution of production incidents to reduce impact on users effectively.
• Continual iterative enhancements guided by real-world usage analytics to improve system stability and performance.

Technical Environment

• Python programming language
• Node.js
• PyTorch framework
• Utilization of OpenAI, Anthropic, and other open-source large language models
• Both SQL and NoSQL databases
• Kubernetes orchestration
• Containerization with Docker

Work Culture

The company operates as a compact, highly skilled, and hands-on team that values collective decision-making, rapid delivery balanced with quality, and learning from practical deployments. New members are expected to bring organization, exercise sound judgment, and work independently to contribute effectively toward delivering a magical AI product to users.

Interview Process

Applicants progressing beyond initial evaluation will be invited for up to four interview rounds conducted virtually or onsite by technical team members. The recruitment process is designed to be transparent and efficient, with prompt decisions following interview completion. Selected candidates will receive an offer to join a team dedicated to delivering AI solutions that provide tangible benefits to billions worldwide.