Backend Engineer, AI Systems

About ActAI

ActAI is dedicated to developing proactive and AI-native applications for over 5 billion users worldwide who currently use basic tools like email, notes, tasks, and calendars without native AI integration. Our goal is to infuse intelligence into everyday workflows such as conversations, errands, and organization, minimizing the need for complex prompting.

Our focus is on building highly reliable applications that maintain persistent context and ensure real-world task completion, significantly reducing inaccuracies common in AI predictions. We aim to help users organize their lives efficiently, enabling them to focus on more meaningful activities.

Role Overview

As a Backend Engineer specializing in AI systems, you will be responsible for designing and managing the inference and orchestration layers facilitating every AI-driven interaction within our product. Your role bridges AI models and end-users, ensuring low latency, accuracy, reliability, and cost-efficiency. You will build and maintain production-grade systems that transform AI model capabilities into fast, stable, and observable APIs accessible across both mobile and desktop platforms.

Primary Responsibilities

• Develop and maintain backend infrastructure that powers AI-driven features in production environments.
• Architect inference pipelines and orchestration mechanisms handling complex multi-step workflows, integrations with external tools, and retries.
• Oversee the complete AI request lifecycle, including routing, caching, batching, streaming, and state management.
• Enhance system performance focusing on optimizing latency, throughput, and operational costs related to model inference and supporting services.
• Ensure system reliability despite unpredictable AI model behavior and dependencies on external systems.
• Implement robust observability solutions for AI platforms encompassing logging, tracing, and debugging of model results and errors.
• Collaborate closely with machine learning and product teams to translate AI model capabilities into reliable and production-ready APIs.

Ideal Qualifications and Experience

• Proven expertise in backend engineering within live production environments.
• Experience managing services that require high throughput and low latency.
• Understanding of AI inference workflows, including knowledge of large language models, embeddings, and multimodal AI.
• Skilled in diagnosing and troubleshooting distributed systems under operational load.
• Proactive approach with a focus on deploying features quickly and learning from production outcomes.

Expected Deliverables

• Backend systems that reliably operate at scale, efficiently managing AI production traffic with minimal delay and maximum throughput.
• Successful execution of complex AI workflows integrating multiple tools and services with comprehensive failure handling and retry mechanisms.
• Stable, clear APIs supporting smooth frontend and machine learning system integration.
• Timely detection, analysis, and mitigation of production issues, minimizing impact on users.
• Continuous improvement of system performance and dependability informed by real-world usage.
• Flexible system designs that accommodate scaling, increasing complexity, and integration of emerging AI features without extensive rework.

Technology Stack

• Python
• Node.js
• PyTorch
• AI models including OpenAI, Anthropic, and open-source large language models (LLMs)
• SQL and NoSQL databases
• Kubernetes container orchestration
• Docker containerization

Work Culture

We believe the most impactful products come from small, highly skilled teams. Our environment values high talent density and hands-on collaboration, with collective decision-making and rapid iteration. We balance shipping quality solutions quickly while fostering continual learning. Successful candidates will demonstrate independent judgment and structured execution to bring innovative AI solutions to users worldwide.

Hiring Process

Candidates who fit the role will undergo a streamlined interview process involving 3 to 4 stages, conducted virtually or onsite. Evaluation is performed by technical team members ensuring transparency and efficiency with prompt hiring decisions. Joining us is more than just employment – it's becoming a part of a mission to deliver practical AI benefits to billions globally.