"About Us Latent AI is a leading AI solutions implementation partner dedicated to helping our diverse clients solve their most complex challenges with cutting-edge technology. Our strength lies in our small agile and expert teams who deliver bespoke AI and machine learning solutions. We foster a culture of collaboration innovation and deep technical excellence. Company Websitehttps//www.reachlatent.com/ The Role Start Date As soon as possibleLocation Remote/Hybrid. Our offices are in Tel Aviv.Employment type Full-time We are looking for a senior full stack engineer to join the team building a production AI-powered customer support platform. It pairs an LLM support agent with a real-time operations and human-review web application and it is already live in front of real customers. The role is hands-on and end to end. You will own the platform from A to Z - the user-facing application the services and data layer behind it the agent runtime the path to production and how the system behaves once it is serving live traffic. The ideal candidate is strong across both frontend and backend and is energized rather than intimidated by a system that combines a live AI agent real-time data third-party dependencies and durable background workflows. This role suits someone proactive who communicates clearly and can work independently as part of a close-knit team. Our Tech Stack Frontend React TypeScript TanStack (Router Query Table Start) Vite Tailwind shadcn/ui Zustand Backend Python FastAPI Pydantic async services webhook-driven integrations Data & Real-Time Convex PostgreSQL object storage (s3/r2) Workflows Temporal for durable long-running execution Internationalization i18next with full Hebrew / RTL support Auth better-auth Testing Vitest Playwright pytest and a containerized full-stack end-to-end harness Monorepo bun and uv workspaces CI/CD & Infrastructure GitHub Actions Docker Kubernetes on GCP Terraform GitOps Observability OpenTelemetry tracing into a centralized observability stack What You'll Do Project & Delivery Team Player You will operate as part of a small team of full stack and Gen AI engineers who build and maintain the platform. Participate in technical discussions reviews and solution design. Full Ownership Own features from planning through the UI services data layer and agent runtime all the way to release and production health - the whole vertical slice not a single layer of it. Client Integrations Own the seams between our platform and our clients' vendor tools - designing them hardening them against real-world failure and diagnosing them when something upstream misbehaves. Proactive Ownership A deep sense of accountability for your solutions the product's quality and its timely delivery. Hands-On Technical Expertise Core Stack Mastery Expert-level hands-on ability across a modern TypeScript and React frontend and a Python backend. You must be able to write production-grade code on both sides. AI Product Engineering Comfort working on a system whose core behavior is probabilistic - prompts retrieval guardrails and evaluation - and the discipline to measure quality rather than assume it. DevOps & CI/CD Solid experience with DevOps principles and CI/CD pipelines. You understand how to maintain and improve a fast reliable and automated path from code to production. Observability Experience implementing or managing logging monitoring and tracing to understand system health debug issues and ensure high availability. Technical Quality & Testing A strong understanding of modern testing strategies (unit integration e2e) and a commitment to shipping high-quality well-tested code. AI Adoption Leveraging AI coding assistants wisely to enhance development efficiency while keeping production-level quality. Who You Are Strong Communicator Fluent in English with the ability to communicate complex technical ideas clearly and confidently to both engineers and clients. Business-Minded Engineer You understand the ""why"" behind the code - connecting technical priorities directly to business outcomes and to the experience of the people the product actually serves. Problem Solver You make recommendations contribute to the team's technical discussions and constantly seek ways to move forward even when dealing with ambiguity. Independent Operator You follow a problem wherever it leads including into systems you did not build and you own the outcome rather than the ticket. Role Requirements Must Minimum of 5+ years of experience as a full-stack engineer specifically working with modern web applications. Expert-level hands-on ability with React and TypeScript. Strong production experience with Python ideally with FastAPI and async services. Hebrew language ability - the product serves Hebrew-speaking customers and being able to read conversations directly is a meaningful edge. Proven experience building and operating third-party API and webhook integrations in production - authentication rate limits retries idempotency and failure handling. Demonstrated ownership of a production system end to end including deployment and live debugging. Working effectively with AI coding agents to enhance productivity. Excellent written and verbal communication skills in English capable of communicating complex technical ideas clearly. Advantage Experience integrating customer support CRM or omnichannel messaging platforms. Experience with real-time or reactive backends. Experience with Convex and relational data modeling. Experience with durable workflow engines like Temploral and Triger. Experience with the TanStack ecosystem or comparable SSR React frameworks. Experience with Tailwind CSS and modern component libraries. Experience with Playwright and containerized multi-service test environments. Experience with Kubernetes Terraform or GitOps deployment workflows. Bonus Points Experience in a consulting or professional services environment. Prior exposure to AI ML or data-driven platforms in production. Why Join Us? This is a unique opportunity to join an engineering team and be part of a growing AI services company - owning an AI product that is already live with real users on a stack we control end to end."