Copy PRINCIPAL AI ENGINEER

Svitla Systems Inc. is looking for a Principal AI engineer for a full-time position (40 hours per week), fully remote in Poland. Our client is a digital marketing company that offers email marketing, social media marketing, event marketing, and online survey tools to help organizations expand their businesses by fostering stronger customer relationships.

This is a hands-on Principal IC role. The client is looking for a technical lead for frontier AI who stays close to the code, brings strong product judgment, and raises the bar for agent systems, evals, harness engineering, and the way the client builds with AI. You will write code, prototype, debug production systems, and establish patterns that other engineers can build on. Currently, 15 AI engineers are organized into three squads.

Requirements

• 10+ years of software engineering experience, including significant technical leadership as a senior individual contributor.
• Proven track record of shipping AI/LLM-powered products to production.
• Deep experience with agent architectures, orchestration, tool calling, and MCP interface design.
• Hands-on experience with Mastra or similar agent/workflow frameworks.
• Strong understanding of workflow design and of when to use deterministic, agentic, or hybrid approaches.
• Strong understanding of context engineering: retrieval, memory, compaction, and long-running state.
• Experience with harness engineering: designing the context, tools, environments, and feedback loops that agents operate in.
• Practical experience building evals, regression tests, traces, and production-quality signals for AI systems.
• Experience building long-running, resumable systems that recover safely from failures.
• Familiarity with AI safety and security concerns, such as permissions, tool authorization, human approval, and untrusted inputs.
• Strong backend and distributed systems experience, ideally with Python and/or TypeScript.
• Strong product sense: the ability to recognize what makes an AI experience useful, intuitive, and trustworthy.
• Demonstrated ability to use AI-native development tools and coding agents effectively.
• Comfort with ambiguity, including investigating unfamiliar systems and unblocking yourself.
• Strong written and verbal communication across distributed teams and time zones.

Responsibilities

• Act as the technical lead for frontier AI across the client's teams, staying current on models, agent architectures, tools, and emerging engineering practices.
• Stay hands-on in the architecture and implementation of production AI systems.
• Design and evolve the client's agent platform, including Mastra, workflows, tools/MCP, context and memory, retrieval, and durable execution.
• Lead the approach to harness engineering, including context, instructions, tools, skills, guardrails, and feedback loops.
• Establish a strong eval discipline across prompts, models, tools, workflows, and end-to-end customer outcomes.
• Work with Product and Design to shape useful AI experiences and quickly prototype ideas.
• Build reusable AI platform capabilities that other engineering teams can adopt.
• Design reliable agent systems that handle retries, partial failures, interruptions, and long-running work.
• Make thoughtful tradeoffs across model quality, latency, cost, safety, and reliability.
• Raise the technical bar through architecture, code reviews, mentoring, pairing, and hands-on implementation.
• Collaborate openly across teams and time zones through clear written proposals, ADRs, prototypes, and technical discussion.
• Help evolve how the client builds software through AI-native engineering practices and coding agents.