Senior Applied AI Scientist: Workflow Agents&Prompt Ownership

Senior Applied AI Scientist: Workflow Agents&Prompt Ownership | Maternity Leave Replacement | 50% Position | Ramat Gan

A leading high-tech company is transforming how businesses drive revenue by turning customer interactions into actionable insights and automated workflows.
The company is developing Workflow Agents, an AI agent-building platform that enables customers to create their own AI agents using data captured across calls, emails, and deals. These agents operate within business workflows, for example, generating meeting briefs automatically. Each agent is built using multiple LLM steps, making prompt design and evaluation essential to the quality of the customer experience.
We are looking for a Senior AI Practitioner to join the Workflow Agents team as its Data Scientist, taking ownership of these LLM steps, including prompt development, evaluation, and quality. Over time, the role will also involve helping engineering teams develop strong capabilities in prompt and skill ownership.

What you’ll do

Workflow Agents: your main focus

• Be the data scientist on the Workflow Agents team and own the feature’s LLM steps: their prompts, evals and quality.
• Help product and engineering design the future of the feature: what it can and cannot do, and which steps and capabilities come next. You will work across the Automations and Custom Agents teams.
• Define how agent quality is measured, offline and in production, and build evals into the product.

Guiding prompt owners: a second focus that grows over time

• Guide engineers who are new to owning prompts and skills: writing prompts, building evals, reading traces.
• Help define how prompt ownership should evolve across Engineering and Data Science. This includes a shared community and playbook for everyone who owns prompts, and ways to automate development and evaluation. Then turn the highest-priority ideas into practical improvements.
• Partner with the agentic framework team to shape tools for experimentation, tracing and evaluation, built for data scientists and prompt owners.

What you bring

• Hands-on experience developing LLM applications and agent workflows using agent frameworks.
• Strong experience designing evals for LLM systems, including LLM-as-judge and regression test sets.
• Strong Python skills and practical experience building evaluations, analyzing traces, and improving prompts based on evidence.
• Experience with tracing and observability tools for LLM apps, such as LangSmith.
• Ability to work independently, make practical tradeoffs, and translate product needs into technical recommendations.

Nice to Have

• Experience mentoring or managing engineers or data scientists, or leading technical work across teams.
• Ability to influence product and engineering decisions without formal authority.

The Details

• Contract for 6–12 months, about 50% time. Freelancers are welcome..
• We are also hiring for a Senior Data Scientist / ML Engineer role. If you fit both, we would love to hear from you.