Founding AI Engineer (RAG / LLM)

To apply, please email your resume to jobs@tradespace.io Overview Location: Remote‑US or San Francisco HQ Compensation: Base $200k – $400 k + Equity Reports to: CTO / Co‑Founder Team: 14 total · 6 engineers Stack: Python · PGVector · OpenAI / Claude / Gemini / Perplexity · Rails glue code About Tradespace Tradespace is an AI-powered intellectual property (IP) management platform on a mission to help innovators protect their groundbreaking ideas. Our AI platform is used by some of the world’s most innovative organizations to develop world-class IP faster and more efficiently. At Tradespace, you’ll be part of a fast-growing Series A startup backed by top investors, working at the intersection of cutting-edge technology and world-changing inventions – from quantum computing and nuclear fusion to life-saving cancer treatments. Why This Role You will be our first dedicated AI hire, owning the Agentic framework we use to discover new innovations and draft world-class patent applications, Green‑field scope, dedicated AI budget and an opportunity to work with a team of elite engineers and IP Attorneys to support some of the most innovative labs and companies in the world. Day‑to‑Day Design multi‑modal embedding stores for text, images & CAD Build high‑recall, low‑latency retrieval APIs at scale Orchestrate LLM‑based agents for patent drafting and analysis Ship Python micro‑services and integrate with our Rails backend Pair with engineers, product and patent attorneys to validate output Experiment with the latest models and roll the best into production What You’ll Own AI Architecture & Roadmap: Model choices, storage strategy, agent workflows Production RAG Services: Embeddings, indexing, retrieval, monitoring Agentic Workflows: Tool‑using LLM chains, human‑in‑the‑loop hand‑offs Quality & Reliability: Testing, observability, statistical debugging Scaling Strategy: Compute budgeting and cost / latency trade‑offs Must‑Have Qualifications 4 + yrs shipping NLP/LLM systems to production (Python) Hands‑on RAG expertise: embeddings, vector DB tuning, latency control Experience debugging LLM outputs with statistical / eval frameworks Ability to explain technical decisions to multidisciplinary stakeholders Bias toward action in zero‑to‑one environments Nice‑to‑Have Multi‑modal embeddings for images or diagrams Exposure to LangChain, AutoGen or similar orchestration tools Familiarity with Rails or other Ruby ecosystems Curiosity about patents and IP workflows Hiring Process (≈ 2 week total) 45‑min technical screen with CTO Paid take‑home exercise (≤ 3 h) — real Tradespace data De-brief with CTO and Product Final interview loop (architecture deep‑dive + culture chat) Benefits & Perks Remote, hybrid or SF‑office flexibility Hardware of choice and home‑office stipend Comprehensive health, dental, vision Unlimited PTO 401(k) Ready to build the AI engine behind tomorrow’s breakthroughs? Submit your resume at: To apply, please email your resume to jobs@tradespace.io