AI Developer - POLY

Aspire Software is looking for a AI Developer to join our team in Lebanon. Here is a little window into our company: Aspire Software operates and manages wholly owned software companies, providing mission-critical solutions across multiple verticals. By implementing industry best practices, Aspire delivers a time sensitive integration process, and the operation of a decentralized model has allowed it to become a hub for creating rapid growth by reinvesting in its portfolio. About the job: This is a builder role for someone obsessed with what agents can actually do in production. You'll design and ship autonomous AI systems that operate inside the Polygon marketplace — agents that understand jewelry listings, assist buyers and sellers, surface intelligence across decades of trade data, and automate workflows that have been manual for forty years. You won't just be wiring up LLM calls. You'll be designing agentic systems end-to-end: tool definitions, planning loops, memory, evals, guardrails, observability, and the product surfaces that put these capabilities in front of real users. What You'll Do Design and build agentic systems that power core marketplace experiences — listing intelligence, semantic and multi-step search, buyer/seller assistants, trade matching, content moderation, and market analytics Architect tool-using agents with reliable planning, execution, and recovery Build the supporting infrastructure: RAG pipelines, vector search, evals, prompt management, traceability, and cost/latency monitoring Partner with product and design to translate agentic capabilities into UX that members actually use and trust Establish patterns and primitives so the rest of the engineering team can ship agent-powered features confidently Stay on the frontier — evaluate new models, frameworks, and techniques and bring the ones that matter into production 3 - 5 years of experience Production experience building agentic systems with LLMs — not just chat wrappers, but agents that plan, use tools, and complete multi-step tasks Deep familiarity with at least one major LLM API (Anthropic, OpenAI, or equivalent), including tool use / function calling, structured outputs, and streaming Experience designing and running evals for non-deterministic systems Strong software engineering fundamentals — TypeScript/Python, modern web stack, API design Pragmatic judgment about when to use agents, when to use simpler pipelines, and when not to use AI at all Nice to have Experience with agent frameworks (LangGraph, CrewAI, etc.) and equally comfortable building from scratch RAG over heterogeneous data — structured listings, images, unstructured text, historical records Multi-modal experience (vision models for jewelry imagery, OCR, image-to-attribute extraction) Background in marketplaces, search, recommendations, or trust & safety Experience with iOS and Android coding & support Experience with Wordpress & front-end SEO Familiarity with the jewelry, luxury, or trading industries Experience deploying agents under real cost, latency, and reliability constraints