Senior AI Engineer - Agentic Products

About Guidde

Guidde is an AI digital adoption platform that serves as the creation and consumption hub for modern organizations. Founded in 2020 by seasoned entrepreneurs, we use generative AI to transform workflows into rich, multi-format documentation in minutes. Our platform helps teams scale knowledge across employees, customers, and even AI systems, turning everyday processes into instant, in-app video guiddes that live where work actually happens.

More than 5,500 organizations worldwide, from fast-growing startups to Fortune 500 enterprises, rely on Guidde to reduce support load, accelerate onboarding, and make knowledge accessible on demand. We’re building the future of how knowledge is created, shared, and experienced.

The Ideal "Guidder" is like our platform - clear, impactful, and focused on delivering value. We blend a startup environment with sports team principles - strategic in planning, collaborative in execution, and obsessed with customer wins. We value doers who listen more than they talk and who show up ready to play their best game. Ready to make your mark on our roster? Go Guidde!

If you are a backend engineer who sees LLMs as a new architecture building block, enjoys balancing deterministic systems with probabilistic AI, and wants to ship real features - not just experiments - then your place is with us!

What will you do?

You build agents that real customers depend on, and you’re the one who knows, with data, whether they’re getting better. At Guidde, agents answer questions over company knowledge, generate documentation autonomously, and guide users inside the apps they work in. You’ll own how they behave in production.

• Build agentic products end to end - multi-step agents with tools, retrieval and memory, from design to production (AskAI, g0, agent-facing MCP surfaces)
• Own quality after launch - golden eval sets, regression gates in CI, production tracing, failure analysis. A release is where measurement starts, not where it ends.
• Engineer context, not just prompts - retrieval, tool design, structured outputs, guardrails, and human-in-the-loop where it matters
• Make trade-offs explicit - quality vs. latency vs. cost, model choice vs. retrieval quality, backed by numbers
• Ship AI-natively - coding agents are your primary dev interface, and you improve the harness that lets them validate their own work
  
  

• 4+ years of production engineering, with strong TypeScript/Node.js
• At least one LLM or agentic feature shipped to external customers and owned after launch - you can tell us what broke and how you found out
• Hands-on experience with evals and LLM observability (LangSmith, Langfuse or similar)
• Solid system design for probabilistic systems: long-running inference, retries, fallbacks, idempotency
• Product sense - you define what “good” means before you build
  
  

• LangGraph, MCP or other agent frameworks in production
• Retrieval at scale (hybrid search, re-ranking)
• Multi-provider LLM routing, fallback and cost tracking
  
  

• Leave Your Mark: Build core AI systems that directly shape how thousands of teams create and share knowledge
• Work Your Way (Hybrid): Enjoy a flexible work arrangement based in Tel Aviv
• Innovate Together: Collaborate with product, backend, and design teams to turn AI capabilities into real customer value
• Fuel Your Growth: Expand your skills across AI systems, backend architecture, and product-driven engineering
• Your Value Matters: Receive a competitive compensation and benefits package