Applied AI Engineer, Founding Team

Tel Aviv (In-office)

Summary:

Early-stage company (seed, led by Greylock) is building a new class of AI system designed to execute complex, real-world workflows end to end. The product moves beyond assistive AI and focuses on systems that can reliably interpret context, make decisions, and take actions across fragmented environments.

These workflows are high-stakes, multi-step, and operate across messy data and existing systems. Success requires combining strong reasoning capabilities with robust execution, reliability, and clear auditability in production settings.

The team is focused on turning recent advances in foundation models into production-grade systems that can operate in real environments, not just controlled demos.

What You’ll Do

• Design and ship agentic AI systems that interpret intent and execute multi-step workflows across real-world environments
• Build planning, tool-use, memory, and recovery mechanisms for agents operating in complex, stateful systems
• Develop execution frameworks that handle ambiguity, partial failures, and changing context
• Create evaluation systems for agent performance, including success metrics, regression testing, and feedback loops
• Improve system reliability through validation layers, guardrails, and well-defined fallback behavior
• Partner closely with engineering and product to bring research ideas into production
• Work with early users to translate real-world workflows into automation coverage

Where You’ll Work in the Stack

• Agent layer: planning, reasoning, and multi-step execution
• Data layer: integrating and structuring fragmented data into usable context
• System layer: orchestration, reliability, and real-world action execution
• Evaluation layer: measurement, monitoring, and continuous improvement

What We’re Looking For

• Experience building agentic AI systems that plan, take actions, and operate across multi-step workflows
• Hands-on work with LLM-based systems beyond demo-level reliability
• Strong understanding of evaluation, including how to measure correctness and improve systems over time
• Solid applied ML fundamentals and ability to move between research and implementation
• Strong engineering skills (Python required; systems experience is a plus)
• Comfort working with ambiguous problems and messy real-world data
• Good judgment around correctness, reliability, and system behavior in production

Signals We’re Especially Excited About

• You’ve built systems that take real actions, not just generate outputs
• You’ve worked on long-running, stateful workflows with failure handling and recovery
• You have strong opinions on architecture and evaluation for agentic systems
• You’ve shipped systems in environments where reliability and correctness matter
• You enjoy turning complex workflows into simple, automated systems

Why This Role

• Tier-1 funded startup
• Frontier problem at the intersection of agentic AI and real-world system execution
• Early team with significant ownership and influence on product and architecture
• Tight feedback loops with early users and real-world deployment
• In-office culture in Tel Aviv with a focus on speed, iteration, and collaboration

About Greylock

Greylock is a 1st-tier, early-stage venture capital firm that partners with exceptional founders at the seed and Series A stages. Our mission is to help realize rare potential — backing category-defining companies such as Figma, Anthropic, Ramp, Rubrik, Airbnb, LinkedIn, Roblox, Dropbox, and Coinbase.

About the Greylock Recruiting Team

As full-time, salaried employees of Greylock, our team provides free candidate referrals and introductions to our active portfolio companies. Combined, we bring over 125 years of in-house recruiting experience across startups and large-scale tech companies