Opportunities in Israel

Full Job Description

Design and build AI agents that plan, use tools/APIs, manage state/memory, and reliably complete multi-step workflows. Own AI features from design through production, including deployment, monitoring, and live‑site reliability, with an eval-first development lifecycle: define success criteria, build evaluation datasets and automated harnesses, and run human-in-the-loop reviews where needed. Develop and maintain prompt, retrieval, and memory strategies (system prompts, few-shot examples, tool schemas, retrieval context) with proper versioning and evaluation coverage. Debug AI behavior using prompt analysis, data inspection, and model/tool-call traces, and translate failure patterns into targeted improvements. Establish and track AI quality metrics (e.g., accuracy, groundedness, relevance, hallucination rate) and integrate them into CI/CD release gates. Optimize runtime performance and economics (token usage, inference cost, latency, caching, model selection/routing, batching) and implement monitoring and continuous improvement loops (online signals, drift detection, structured user feedback). Partner with product, design, and domain stakeholders to define use cases, acceptance criteria, and rollout plans for AI features. Live site responsibility You have at least 7+ years professional software development with at least 4+ years of software engineering experience in the AI space (e.g., building and shipping AI/ML or GenAI features in production) You have proven experience with building AI agents Hands-on experience with evaluation methodologies and integrating quality standards/guardrails into delivery. Proficiency in Python and/or C#, with experience using REST APIs and SDKs. Deep understanding of AI system design, including ML fundamentals, Generative AI concepts, and cloud-native architectures. Bachelor's degree in computer science, Engineering, or equivalent practical experience. Strong context engineering and debugging skills across prompts, tool schemas, retrieval pipelines, and model behavior (not only code-level debugging). Ability to work effectively with non-deterministic/probabilistic systems and design reliability despite variable outputs. Proficiency in software engineering fundamentals (APIs, data structures, CI/CD, observability), applied to AI systems. Azure stack: Experience shipping production-grade AI features (LLMs and/or classical ML), on Azure, with measurable quality metrics. Experience with LLM observability/tracing and eval tooling, including building internal quality gates and optimizing inference cost/latency in real-time systems. Experience with retrieval systems (indexing, chunking strategies, reranking) and grounding techniques. Ability to govern AI outputs: define quality standards and guardrails, apply responsible AI practices, and put monitoring/evaluation in place to maintain reliability over time. Experience in driving innovation and creating new initiatives from the ground up. Proven ability to work independently, own large problem spaces, and collaborate across disciplines. Ability to collaborate in a fast‑paced