I am an Electrical Engineer turned Software Engineer, with experience in:
- Sensing, Signal Processing: Computer vision, sensor signal processing (inertial, PIR, acoustic, etc.), sensor-based behavior recognition, and more.
- Data Analysis: Processing large datasets from internal logs, from 3rd party sources, from distributed hardware sensors, etc.
- Systems: Distributed, low-power, and resource-constrained embedded systems.
- Visualizations: Skilled at conveying high-level information with interactive visualizations.
- Web development: Full stack engineer, developed numerous web apps, management consoles and dashboards.
- Android development: Research-oriented and prototype apps on mobile phones.
I will let you know more info when I can :)
Founding member of the probject, and tech lead for the web team.
All I can say is we were working on something AI–adjacent, involving something crowd-sourcing–like.
Tech lead for the project. Worked with the team on system architecture, web frontend engineering, backend engineering, product definition, and even some web / Android UX design. We were a small team of 15 or so people, and everyone wore lots of hats.
Optimization of datacenter capacity placement around the world, with the goal of minimizing quantities such as latency, cost, and power consumption, and maximizing revenue. This included analysis of large-scale datasets such as web traffic from different countries/languages/devices as well as worldwide socio-economic trends, to build models of how such quantities vary over space and time. Using these models, we produced recommendations for datacenter capacity placement around the world for the next few years. I also created tools to allow other engineers to quickly use our models, visualize the many inputs, and interact with our results. We worked towards automating the entire datacenter capacity optimization problem.
Working in a tightly knit team, at times collaborating on our various sub-projects, and at other times taking a leading position in them; Coordinating with other groups in order to facilitate greater understanding of our results, as well as greater sharing of our algorithms and codebase; Producing written reports, arranging regular meetings, and making presentations to a wide audience of interested parties.
Worked in the EU-funded CHOReOS project to build a system for service choreographies in the Internet of Things (IoT). Designed the high-level architecture for a data-fusion middleware that would allow non-experts to query the IoT and automatically produce physical inferences and actuations. Started development of this middleware, building, as a first step, a sensor-management service for Android- and Linux-based “things”, called ThingLib.
Driving the design of our data-fusion middleware for the IoT; Collaborating with other universities and research institutes involved in CHOReOS; Co-supervising PhD and Masters students.
Devised a person-tracking and identification system that employs cameras and wearable sensors to identify in a video each each person carrying an inertial-measurement unit (such as a modern cellphone). Developed a lightweight tracking algorithm for real-time operation in resource-constrained embedded systems. Developed drivers, middleware, and a web-based console to operate and remotely manage a network of camera nodes. Investigated different methods and sensing modalities for person detection and activity recognition.
Responsible for driving the person-tracking and indentification project, developing an end-to-end human-sensing solution; Building and maintaining remote deployments of sensor nodes in real homes for multiple months without physical access; Supervising undergraduate research projects and graduate-level class projects.
|Yale University||Ph.D.||Electrical Engineering||Dec. 2010|
|Yale University||M.S., M.Phil.||Electrical Engineering||May 2007|
|Johns Hopkins University||M.S.E.||Electrical Engineering||May 2005|
|Johns Hopkins University||B.A.||Mathematics||May 2003|
|Johns Hopkins University||B.S.||Electrical Engineering||May 2003|
Published over 20 peer-reviewed articles. An abridged list follows. The full list is available at thiagot.com/publications.html.
- IEEE Journal of Selected Topics in Signal Processing 2008
- Proceedings of the IEEE 2008
- ACM Transactions on Sensor Networks 2009
- Invited paper at ServiceWave 2011, LNCS 6994
- ACM International Conference on Ubiquitous Computing (UbiComp) 20101 of 5 Nominees for the Best Paper Award
- IEEE/ACM International Conference on Distributed Smart Cameras (ICDSC) 2009
- ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN) 2006
- Subject Sensing in an Environment (pending)WIPO publication number WO/2013/003860
Graded on a most unscientific scale.
Programming and Hardware Design Languages
- Advanced: R, SQL
- Intermediate: C++, C#, VHDL
Markup and Style Languages
- Fluent: CSS, HTML, LaTeX, XHTML
Tools, Libraries, Frameworks and Environments
- Fluent: AngularJS, Closure, d3.js, Django, jQuery, Pylab
- Advanced: OpenCV, Xilinx ISE
- Intermediate: Cadence, HSPICE simulator
- Fluent: Portuguese (native), English
- Intermediate: Spanish, French