Build intelligent AI workflows to automate document-heavy back-office and front-office processes across departments:
Architect Retrieval-Augmented Generation (RAG) systems that allow staff across all departments to query internal SOPs, technical standards (e.g., IEC/BS), product catalogs, and historical business records via natural language.
Connect Generative AI pipelines and APIs with existing enterprise systems (ERP, CRM, HRMS, Microsoft 365, Google Workspace, and internal file shares) to trigger hands-free cross-platform workflows.
Develop autonomous multi-step agents (using tools like LangChain, n8n, Flowise, or custom Python microservices) that read emails, extract structured data, update databases, and draft responses.
Implement prompt guardrails, data validation checks, and role-based access control to ensure outputs are reliable, hallucinations are minimized, and sensitive commercial and employee data remains secure.
Required Qualifications & Skills
Bachelor’s degree in Computer Science, Artificial Intelligence, Data Science, Software Engineering, or related technical field.
2+ years of experience building applications with Large Language Models (LLMs), prompt engineering, and RAG architectures using vector databases (Pinecone, Chroma, FAISS, Milvus).
Strong proficiency in Python, RESTful API design (FastAPI/Flask), webhooks, and JSON data manipulation.
Proven experience with Document AI, Intelligent Document Processing (IDP), multi-modal models, or OCR (e.g., extracting structured tables and text from PDFs, POs, and invoices).
Strong communication skills to liaise with non-technical department heads (Purchasing, Finance, HR, Sales, Production) to map processes into automated workflows.
Practical exposure to no-code/low-code integration platforms (n8n, Make, Power Automate) alongside code-based AI frameworks.
Experience deploying local/on-premise open-source LLMs (via Ollama, vLLM) for data privacy and internal-only usage.
Familiarity with ERP workflows, purchase order lifecycles, or supply chain documentation standards.