Applied AI Engineer

The Job

We are looking for an experienced and technically strong Applied AI Engineer to join us and take full ownership of the design, development, and deployment of LLM- and agent-based systems supporting business processes within a client organization.

This is a highly independent role. You will sit with the business stakeholders, understand how their processes actually work, identify where AI genuinely creates value and where it does not, and build and ship the systems yourself. Technical and architectural guidance is provided by Galil's Chief AI Officer, but day to day you are the AI function inside that organization.

Working with AI coding tools such as Claude Code and Codex is a core part of the role. We expect that — and we equally expect you to understand and take full responsibility for everything they produce.

Responsibilities

• Working directly with business stakeholders to map existing processes, elicit requirements, and identify high-ROI automation opportunities — including the judgment to say when something is not worth building.
• Designing, developing, and maintaining production systems built on LLMs, RAG, and agent architectures.
• Integrating these systems with the organization's existing platforms, data sources, and authentication infrastructure.
• Building the internal-facing interfaces users actually work in.
• Defining success metrics and evaluation methods, and proving the systems work before and after they go live.
• Full ownership of deployment, monitoring, cost management, and continuous improvement.
• Communicating technical concepts clearly to both technical and non-technical audiences.
• Writing the specifications and documentation that let others understand and maintain what has been built.

Requirements

• At least 5 years of hands-on Python development experience.
• Bachelor's degree in Computer Science, Software Engineering, or a related field from a recognized institution
• Proven experience owning a system end to end: design, implementation, deployment, and maintenance after go-live.
• Practical understanding of how language models behave and why they fail: context management, prompt design, hallucination, and non-determinism — enough to debug behavior rather than guess at it
• In-depth RAG experience: chunking, embeddings, retrieval, re-ranking, and evaluation — along with the judgment to recognize when RAG is the wrong tool
• Familiarity with agent architectures: tool use, MCP, orchestration, human-in-the-loop design, and common failure modes
• Experience defining success metrics, building evaluation sets, and running regression tests when prompts or models change
• Demonstrated hands-on experience with AI coding tools in real projects, together with critical review of the code they produce
• Proven experience defining and implementing RESTful APIs and integrating third-party interfaces
• Experience working with databases: SQL, NoSQL, and vector databases
• Experience with cloud platforms: Azure or AWS, including Docker and CI/CD
• Front-end knowledge: HTML, CSS, JavaScript and a modern framework (React or equivalent) — enough to build and maintain internal-facing tools without a dedicated front-end developer
• Practical security understanding: correct handling of secrets and credentials (never in code, prompts, or logs), enforcement of access control at the retrieval layer so a RAG system never returns data the requester is not permitted to see, and awareness of LLM-specific risks such as prompt injection and data exfiltration through tool calls
• Sound judgment on what data may be sent to third-party model providers, and on retention, PII, and audit logging
• Full command of Hebrew and English, including excellent written and verbal communication in both languages
• Proven ability to work independently with minimal supervision — arriving with a plan rather than a list of questions, while knowing when a decision should be escalated
• Strong client-facing skills and experience working directly with business stakeholders, including translating vague requests into clear specifications
• High self-motivation, alongside excellent time management and organizational skills

You do not need experience training models. You do need to understand them well enough to build reliable systems on top of them.

Advantages

• Significant advantage: prior experience deploying AI systems inside organizations and managing organizational change, not only building
• Significant advantage: experience in the client's industry
• Background in data engineering or analytics
• Prior experience in client-facing technical roles or as an embedded engineer
• Knowledge of prompt engineering and AI-assisted development workflows

About Us:

We are Galil Software, Israel’s premier onshore outsourcing option for R&D and software services.

Located in Nazareth Industrial Park, Galil Software is Israel’s leading provider of software R&D Extension, QA Automation, DevOps services and Web and Mobile Projects.

Since its founding in 2008, Galil Software has successfully established a proven track record for increasing the productivity and efficiency of R&D operations. Managed by Hi-Tech executives with experience in running global organizations, together with more than 200 software engineers – Galil Software offers an end-to-end solution to R&D and customer success departments.

Some of the leading Hi-Tech companies that Galil Software proudly lists among its customers include Western Digital, Radware, Imperva, Allot, AT&T, Taboola, Via and more…