Role Overview Build and deploy an end-to-end AI-driven SEO agent platform. You will own everything from prompt-engineering and RAG pipelines to CMS integration, staging environments, CI/CD, and a minimal frontend or plugin for demos. Key Responsibilities Design and implement LLM-powered agents (prompt flows, retrieval-augmented generation, embedding stores) using frameworks like LangChain, LlamaIndex (formerly GPT Index), or similar. Prototype and integrate vector-based retrieval with Pinecone, Weaviate, or Milvus. Build and secure REST APIs for WordPress, Shopify, headless CMS or React/Node.js platforms. Containerize services with Docker; orchestrate staging via Kubernetes or Terraform. Implement CI/CD pipelines (GitHub Actions, GitLab CI) with rollback and health-check gates. Scaffold lightweight frontend components or CMS plugins (React/Vue) to showcase agent capabilities. Automate SEO tasks on-page (metadata, schema injection), internal linking, and 404 recovery via code. Collaborate with prompt engineers and QA to validate agent outputs before production push. Required Experience 2+ years of software engineering with 2+ years in AI/ML or NLP projects. 1-3 years Python (FastAPI, Django) for API and agent development. Hands-on with LLM models (OpenAI GPT-4, Claude, or self-hosted LLaMA/Meta’s models) and fine-tuning pipelines. Experience with prompt frameworks: LangChain, LlamaIndex, Haystack, or equivalent. Proven RAG (Retrieval-Augmented Generation) implementation experience. Proficiency in Node.js and React (or Vue) for plugin/UI development. Strong background in Docker, Kubernetes (or Terraform), and CI/CD tools. Familiarity with CMS integration: WordPress REST API, Shopify Admin API, or headless CMS APIs. Solid prompt-engineering skills: few-shot, chain-of-thought, retrieval-augmented prompts. Knowledge of SEO best practices: metadata, JSON-LD schema, internal linking, and redirect management. Experience writing automated tests and health-check scripts for staging validations. Nice-to-Have Prior work on SEO automation, site-audit tooling, or custom CMS plugins. Experience deploying or fine-tuning open-source LLMs (e.g., LLaMA, Falcon, MPT). Exposure to Haystack, Retrieval QA stacks, or vector pipeline orchestration. Familiarity with monitoring/logging stacks (Prometheus, ELK, Grafana). Basic understanding of frontend styling (Tailwind, Material UI) for rapid prototyping. Apply if you can take full ownership of AI agent development and deployment, leveraging LangChain, style frameworks, and a range of LLMs, to deliver a working MVP in 4–6 weeks. Show more Show less