LLM Application Engineer

About ActAI

ActAI is dedicated to creating intelligent, easy-to-use applications for over 5 billion users who currently utilize basic tools such as email, notes, tasks, and calendars without AI integration. Our mission centers on delivering proactive AI-powered experiences that require minimal to no user prompting, improving workflows, organization, and daily tasks with high reliability and reduced hallucinations.

Role Overview

As an LLM Application Engineer, you will be responsible for developing the core intelligence layer behind ActAI's AI solutions. This role combines advanced language models, software engineering expertise, and product-focused development to design agent workflows, enhance AI performance, and deliver dependable user experiences. You'll lead end-to-end problem-solving, from comprehending user requirements to deploying, evaluating, and refining AI systems in production.

Key Responsibilities

• Create and deploy LLM-driven applications and agent workflows.
• Architect systems that support reasoning, planning, memory, tool usage, and multi-step task execution.
• Develop robust orchestration pipelines that convert uncertain model outputs into consistent, traceable, and secure actions.
• Integrate language models with APIs, databases, search functionalities, and both internal and external tools.
• Implement techniques such as advanced prompting, context optimization, structured output generation, and tool invocation to boost AI behavior.
• Construct evaluation frameworks and datasets to assess AI quality, reliability, and detect regressions.
• Diagnose issues across layers including model behavior, prompts, orchestration logic, backend services, and user interface.
• Optimize AI systems for performance metrics including quality, responsiveness, and operational costs.
• Collaborate closely with product and engineering teams to convert complex requirements into functional AI solutions.
• Promote best practices for production including observability, tracing, experimentation, evaluation, and ongoing improvement.

Technology Stack

• Python programming language
• LLM APIs and compatible model providers, inclusive of OpenAI and open-weight models
• Agent orchestration frameworks
• Vector databases and retrieval systems
• Backend services, APIs, and distributed computing platforms
• Machine learning frameworks such as PyTorch and JAX

Preferred Qualifications

• Solid foundation in software engineering and experience developing AI-powered products
• Practical exposure to large language models, generative AI, or agent system architectures
• Expertise in designing prompts, workflows, evaluation methods, and AI behavior tuning
• Ability to produce clean, maintainable production-level code
• Versatility in working across technology layers—from models to systems to product implementations
• Strong analytical and problem-solving capabilities in dynamic and uncertain settings
• Focus on rapid deployment, iterative development, and continuous system enhancement

Expected Outcomes

• Accelerated release of AI features that measurably improve user experience
• Creation of reliable, scalable, and maintainable LLM-based workflows
• Enhancement of AI quality through systematic testing, experimentation, and refinement
• Development of AI workflows that are more predictable, efficient, and cost-effective over time
• Translation of complex AI functionalities into simple and user-friendly applications