Senior AI Engineer - Sticklight

We are seeking an exceptional Senior AI Engineer to join our core AI team, working on Sticklight, a revolutionary AI coding agent inside a browser-based IDE with live preview. Users describe what they want, the agent builds it in real time, for anyone building full-stack web apps.

In this role, you'll own product features from start to finish. Not just the code, but also understanding what users need, designing the solution, building it, shipping it, watching how it performs. There's product and design support, but we want someone who also drives things forward without waiting to be told what to build. If you care about product outcomes and live in the world of agents, prompts, evals, and tool orchestration - keep reading.

Your mission is to apply cutting-edge AI technologies, architect and deploy solutions that are high-performing, and being at the leading LLM based technological exploration to utilize a diverse array of tools and advanced technologies that creates a seamless experience for our customers.

If you’re enthusiastic about AI and you breathe Agents, MCP, Prompt Engineering, high pace delivery, and all the goods of the new world, your place is with us.

What do we do

Elementor is where web creation happens for real.

We are a multi-product platform powering over 13% of the internet.

Our products give millions of web creators: developers, designers, and marketers - everything they need to build, scale, and succeed.

From Israel to more than 160 countries, a new website goes live with Elementor every 10 seconds.

How do we do it

By staying AI-first, data-driven, and human-led.

We build real products that deliver real value, and we do it together.

At Elementor, creativity, fast feedback, and open collaboration shape everything we do.

You will find good energy and people who genuinely care about the product, the users, and each other.

It is a place to grow, make an impact, and help shape the future of the web.

Responsibilities:

• Own features end-to-end, from the user problem through to production monitoring
• Build and evolve the AI agent: prompts, tool calling, context management, multi-step reasoning
• Evaluate and improve LLM output quality, speed, and cost
• Maintain production across a microservices architecture with real user traffic
• Test new models and techniques. Ship the ones that work
• Work with product and design to figure out where AI actually helps
• Contribute to a small team where everyone builds everything

Requirements:

• 5+ years TypeScript full-stack (Node.js + React)
• 2+ years building LLM applications in production. Evals, fine-tuning, agentic workflows
• Experience with LLM evaluation: building eval sets, measuring quality, iterating based on data
• Experience building with AI-native tools and agentic workflows (ie. prompt engineering, tool use, plugins, automation frameworks)
• Worked with most of: PostgreSQL, Kafka, Redis, Docker, Kubernetes. You don't need all of them day one
• You take ownership and deliver independently
• You think about users before you think about architecture
• Advantage: Background in DS/ML with a transition into Software Engineering
• Advantage: Experience with tracing and observability (Langfuse, Sentry, Grafana)

What makes this role different:

• Small team, top-tier developers, everyone full-stack, everyone touches everything. As the team grows so does your scope
• The engineers on this team have shipped production AI systems before. You'll learn from them and they'll learn from you
• You'll shape the product directly. Example: our eval framework, agent tool system, and sandbox architecture were all designed by the engineers who built them, not handed down from a roadmap
• Elementor is behind this. Real company, real resources. But the team runs like a startup - small, fast, and independent

Here at Elementor, we encourage our employees to express their personality and identity (whether gender, ethnic, religious, or sexual), and we ensure fairness and equal opportunities. Our hybrid work model is designed to support performance while respecting individual needs.