Job Description

Passionate about food? Eager to join a dynamic SaaS startup shaping the food and beverage industry? Tastewise is just the place for you.

Tastewise: The GenAI-powered consumer data platform revolutionizing product strategies and driving success for major brands like Nestle, Pepsi, and Campbell’s. With insights from billions of data points, we empower our clients to create, market, and sell their products faster and more successfully.

Join this $10 trillion industry, where creativity and an entrepreneurial mindset drive daily success and reach the homes of billions of people in an industry where change isn't just a recipe but a way of life.

Let's shape the future of food together.

About the role

We’re looking for a product-minded AI Engineer with a strong software engineering background, who knows how to turn advanced AI capabilities — including agentic systems and MCP — into production-ready features.

This is not a research role and not an experimentation sandbox. It’s a hands-on engineering role focused on ownership, systems thinking, and shipping AI as a core part of a real product with users, data, and real-world constraints.

What you’ll do:

• Own AI features end-to-end — from defining the business problem to production deployment
• Build LLM-based systems, including: RAG and Retrieval / Semantic Search systems & Agentic Solutions: Agents, Tool Use, Orchestration, Memory, and Control Flows
• Design and implement Agentic Systems using Model Context Protocol (MCP) for structured tool access, context management, and modular agent interactions
• Build systems where models make decisions, invoke tools, and perform actions in a controlled and observable way
• Integrate AI into existing production systems — APIs, async jobs, data pipelines, monitoring
• Make architectural decisions around models, providers, and Cost / Latency / Reliability tradeoffs
• Ship production-grade solutions with logging, retries, testing, and observability

How we think about AI:

• We build Systems, not just prompts
• We’re pragmatic — not every problem needs an agent
• We care deeply about Quality, Cost, and Latency
• We actively use AI to write better code, faster

Who this role is for:

• Engineers who made a real transition into AI through building and shipping
• People who enjoy ambiguity, iteration, and system-level problem solving
• Those who want to work on AI that is part of the Core Product, not a side experiment

Requirements

• 4+ years of experience as a Full-stack or Backend Engineer
• At least 2 years of hands-on experience delivering AI/LLM-powered features into production, or equivalent real-world project scope.
• Thinks like a software engineer first — AI is a tool, not the goal
• Practical experience with: LLM APIs (OpenAI, Claude, Gemini), Prompting, Evaluation, and Debugging of LLM outputs, Building Retrieval / RAG Pipelines
• Experience or deep understanding of Agentic Architectures, including: Task Decomposition, Tool Invocation, State and Memory Management
• Hands-on familiarity with Model Context Protocol (MCP) or similar protocols/architectures for managing context and tools for LLM agents
• Production experience with Python, APIs, Docker, Cloud infrastructure, and CI/CD
• Comfortable working with real, messy data: SQL, Data Processing, and Pipelines

Please note that this role is on a hybrid model of 4 days/week in our Tel-Aviv office.