Principal Data Scientist (AI)

Location: R&D, Ramat Gan, Israel; onsite

Company

hirify.global is a trust, safety, and security company focused on AI lifecycle protection, including model hardening, red-teaming, runtime guardrails, and drift detection for frontier model labs, enterprises, and user-generated content platforms.

What you will do

• Own the GenAI safety roadmap covering guardrails, LLM-as-judge adjudication, prompt-injection and jailbreak detection, policy authoring, and automated red-teaming.
• Choose, fine-tune, evaluate, and retire model families and providers before deployment to customer traffic.
• Design benchmarks, datasets, LLM-jury labeling protocols, multilingual, multi-turn, and multimodal evaluations, including calibration, threshold selection, and error analysis.
• Lead high-stakes GenAI projects from framing and prototyping through offline evaluation, shadow traffic, and production rollout.
• Mentor data scientists and ML engineers through design, experimentation, and model reviews.
• Translate model behavior into product and customer outcomes across accuracy, latency, cost, and enterprise readiness.

Requirements

• Proven technical leadership as a tech lead, senior data scientist, or ML engineer, taking multi-person, multi-quarter ML initiatives from framing to production without formal authority.
• Strong Python skills and written communication for design and evaluation proposals.
• Deep understanding of how LLMs are trained, adapted, and served, including prompting, fine-tuning, RAG, and agentic patterns.
• Ability to measure and improve model quality, latency, and cost, including understanding where different LLM approaches fail.
• Deep, daily fluency with an AI coding assistant such as Claude Code, Cursor, or Codex, with judgment to establish effective team usage.

Culture & Benefits

• Full-time R&D role based in Ramat Gan, Israel.
• Hands-on individual-contributor leadership with ownership of technical direction and architecture.
• Work focused on protecting communicative technologies and safeguarding users across AI and technology platforms.