I am a Senior R&D Software Engineer and a Researcher in Computer Science.I am interested / trained / have worked in:
- Computer Science research, in particular databases and data management, with a focus on data search and access optimization
- Software Engineering, in particular indexing / search software, as well as infrastructure, middleware layers and cross-platform porting
- Teaching, with a strong interest towards enforcing connections between the software industry and academia
Welcome to my website! I am currently a post-doctoral researcher at INRIA / UCSD. I have defended on Oct. 24th, 2014 my PhD Thesis Query rewriting using views: a theoretical and practical perspective. Prior to my PhD I have worked for several years in software engineering (Dassault Systemes, Amazon). Here is my short CV.
My PhD and post-doc research projects, as well as an important part of my R&D work in the software industry, have been oriented towards data search and access optimization, speedup and scalability. I have also worked on (and hold a long term interest in) pattern recognition and similarity search. In the search-and-access speedup area, I have in particular focused on query rewriting (for optimization purposes and targeting practical performance) and on indexing strategies (mostly for multi-dimensional data). Here are some of the topics I've been working on:
- Relational conjunctive queries reformulation under constraints. My SIGMOD '14 paper describes the Provenance-Aware Chase & Backchase, an efficient conjunctive query reformulation algorithm, as well as the main scientific result of my thesis. You can also browse the code here.
- Multi-dimensional indexing. I have investigated efficient multi-dimensional indexing in the context of the ACM SIGMOD Programming Contest 2012. These SIGMOD '12 slides detail my solution, winner of the second prize in the contest. The code is available here.
- XPath queries rewriting with views. My PhD work on efficient techniques for XPath rewriting is part of a paper currently under second revision for TCS. An early version of this paper is available here. You can also take a look at the rewriting code.
- A scalable and efficient multi-model and multi-storage architecture. I am currently working on the INRIA / UCSD project Estocada, targetting the efficient, scalable and adaptive usage of multiple, heterogenous storage systems. This CIDR '15 paper describes the Estocada system's goals and global architecture.