AI Engineer at Jobgether – Israel, Amazonas

This position is posted by Jobgether on behalf of a partner company. We are currently looking for an AI Engineer in Israel.

This role sits at the core of building and scaling advanced LLM-powered products in a fast-moving, product-driven environment. You will design and optimize production-grade AI systems that directly impact real consumer applications across health, education, and entertainment domains. The position blends deep engineering with applied AI, focusing on prompt systems, evaluation pipelines, and model performance optimization. You will work with modern TypeScript/Node.js architectures while integrating cutting-edge LLM tooling and observability frameworks. A strong emphasis is placed on experimentation, data-driven iteration, and cost-efficient AI delivery. You will take ownership of AI features from early experimentation through to production deployment. The environment is highly dynamic, innovation-focused, and geared toward rapid iteration and measurable impact.

You will be responsible for building, optimizing, and scaling LLM-driven systems, ensuring they are reliable, cost-efficient, and production-ready while continuously improving model performance through structured experimentation and evaluation.

• Design advanced prompt engineering systems with dynamic logic, contextual injection, and reusable templates to maximize model reasoning quality and output consistency.
• Implement structured output frameworks (JSON schemas, function calling, Zod validation) to ensure AI responses integrate seamlessly into product logic.
• Build and maintain evaluation pipelines using tools like Langfuse to measure response quality, track performance, and enable real-time feedback loops.
• Conduct deep tracing and debugging of LLM chains to identify inefficiencies, optimize latency, and reduce operational costs.
• Run AI A/B testing across multiple models via platforms like OpenRouter, analyzing performance using quantitative metrics and benchmarking systems.
• Develop scoring and analytics systems to evaluate reasoning chains, detect hallucinations, and improve output reliability.
• Reassess and fine-tune models as needed, ensuring continuous improvement aligned with evolving architectures and business needs.

This role requires strong backend engineering expertise combined with hands-on experience in applied AI systems, LLM workflows, and data-driven experimentation.

• 5+ years of experience with Node.js and Next.js, building scalable backend systems and handling complex data flows.
• Strong experience in prompt engineering, especially dynamic and context-aware LLM prompting.
• Hands-on experience with LLM observability tools such as Langfuse or similar platforms.
• Experience using OpenRouter or unified LLM APIs, including model selection, rate limit handling, and cost optimization.
• Familiarity with evaluation frameworks such as RAGAS or custom LLM-as-a-judge systems.
• Strong analytical mindset with the ability to translate logs and model outputs into actionable insights.
• Experience with structured AI workflows, debugging, and iterative product improvement cycles.
• Nice to have: experience with fine-tuning models, RAG architectures, and basic Python for AI/data tasks.
• Strong communication skills and a highly iterative, experiment-driven working style.

• Fully remote working environment with flexibility to work from anywhere.
• Unlimited paid time off to support work-life balance and personal wellbeing.
• Paid national holidays depending on your location.
• Company-provided MacBook to support high-performance work.
• Flexible independent contractor agreement with autonomy and entrepreneurial flexibility.
• Opportunity to work on cutting-edge AI systems at scale in a high-growth environment.
• Fast-paced culture with strong focus on ownership, impact, and continuous learning.