Armen Inants

I am an AI researcher, engineer, technology leader, and entrepreneur with 15+ years of experience building enterprise-grade software and research-driven pilot projects. I connect academic research and industrial deployment, from Document AI platforms used by Fortune 500 companies to large-scale retrieval-augmented generation (RAG) systems serving millions of documents. My core expertise is in integrating knowledge-driven systems with generative AI to address complex enterprise problems.

I received my PhD in Mathematics and Computer Science from INRIA and the University of Grenoble Alpes. My thesis contributed to common-sense spatial reasoning and ontology matching, and established a link between these previously unrelated areas of AI. It is also cited in A Guided Tour of Artificial Intelligence Research. Volume I: Knowledge Representation, Reasoning and Learning.

I introduced and taught courses on Artificial Intelligence and Decision Support (CS346) and Knowledge Representation (CS347) at the American University of Armenia.

I worked as a postdoctoral researcher at Grenoble Institute of Technology, where I developed an engine for combinatorial auctions in Haskell, enabling research teams to run auction simulations for AI agents.

Previously, as Principal Symbolic AI Research Engineer at Morningstar, I led an AI project for automating predictive analytics, successfully combining symbolic and generative AI within a RAG system grounded in millions of documents.

Currently, I am building a document engine that bridges the gap between human-centric documents and AI-ready data.

Publications

  1. (). . Artif. Intell.
  2. (). . In The Semantic Web - {ISWC} 2016 - 15th International Semantic Web Conference, Kobe, Japan, October 17-21, 2016, Proceedings, Part {I}, pages 360--375.
  3. (). . PhD thesis. Grenoble Alpes University, France.
  4. (). . In The Semantic Web - {ISWC} 2015 - 14th International Semantic Web Conference, Bethlehem, PA, USA, October 11-15, 2015, Proceedings, Part {I}, pages 253--268.

PhD Thesis

I conducted my PhD research in the EXMO team of INRIA and LIG under the supervision of Dr. Jérôme Euzenat. The full thesis can be downloaded here or from the .

My PhD work focused on helping symbolic AI perform common-sense spatial reasoning in a more "human" and flexible way, especially in situations where different kinds of spatial things coexist. Earlier systems for qualitative reasoning (using ideas like "before/after" or "inside/next to" instead of precise numbers) usually assumed a single, uniform universe—only regions on a map, or only time intervals. I developed a modular framework that allows several such reasoning systems to be combined and linked, so an AI system can reason jointly about locations, routes, and landmarks, without breaking the underlying logic.

Qualitative calculi with heterogeneous universes. (Calculs qualitatifs avec des univers hétérogènes)Calculs qualitatifs avec des univers hétérogènesPhDGrenoble Alpes University

Last updated: