Kevin Bals

AI Engineer specialized in production-grade Generative AI systems, RAG architectures and cloud-native backend platforms for industrial clients. • Designed and developed GenAI backend services using Python, FastAPI and Google Cloud Platform. • Built RAG and document intelligence systems enabling users to query complex technical documentation with sourced, citation-backed answers. • Implemented AI agents and multi-step LLM workflows using LangGraph, Google ADK, Vertex AI, Gemini and Azure OpenAI. • Developed hybrid retrieval pipelines combining vector search, BM25, pgvector, PostgreSQL and document hierarchy boosting. • Built document ingestion workflows for PDF, PPTX and XLSX files using PyMuPDF, OCR, LLM/Vision extraction and embeddings. • Deployed cloud-native AI services using Cloud Run, Cloud SQL, Firestore, GCS, Pub/Sub, Docker, Terraform and CI/CD pipelines. • Contributed to hallucination mitigation, retrieval transparency, RAG evaluation and production reliability. • Collaborated with product owners, functional experts and engineering teams to turn complex business needs into scalable AI products.

Worked on industrial innovation projects involving Extended Reality, autonomous inspection, technical document digitalization and computer vision. • Developed software prototypes using Unity 3D, C#, C++ and Python. • Contributed to computer vision and XR workflows for aerospace and industrial environments. • Collaborated with engineering and functional teams to explore innovative solutions for operational use cases.