Engineer Management

SANKALITA SAHA

*** ********* *** ****: 240-***-****

Redwood City, CA 94065 Email: *********@*****.***

SUMMARY OF QUALIFICATIONS

Strong expertise in machine learning and data mining with special focus on regression

techniques, Bayesian learning and decision trees

Extensive experience in design methodologies for embedded signal processing systems

Extensive experience with a wide variety of low-level and high-level programming languages

and concurrent programming

Excellent analytical and problem solving skills

Goal-driven self-starter with proven experience of taking new research/product ideas from

concept to implementation

Excellent team player with ability to collaborate across diverse groups

Extensive experience with communication of technical work through conference publications

and presentations (More than 35 publications in peer-reviewed conferences and journals)

WORK EXPERIENCE

Principal Algorithms Engineer, Jawbone Inc., June 2012-current

Research, development and implementation of a wide range of machine learning and signal processing algorithms

for a sensor band worn at wrist

Developed and implemented complex features for training step classifiers based on in-

depth signal analysis to improve performance under resource limitations

Lead engineer for developing and implementing a sleep classification system

Performed detailed feature analysis, and training of a sleep classifier using a combination

of decision trees and logistic regression from offline repositories of accelerometer data.

The trained classifier was validated using PSG data and integrated into a real-time

activity tracker to detect sleep stages

Developed detailed sleep heuristics based user-friendly display of classification results

Lead engineer for exploring and evaluating feasibility of new product features

Designed and implemented a gesture recognition and tracking system based on wrist

motion using particle filters

Developed and implemented decision tree based activity classification system using

offline repository of accelerometer data that can distinguish between sport activities

such as tennis, swimming, golf etc

Lead engineer for design and implementation of heart rate and respiration rate estimation

system from bio-impedance signals collected at wrist

Designed and implemented a signal chain to detect and track heart rate and respiration

from noisy bio-impedance data. Challenges involved handling measurement noise,

separation of heartbeat and respiration signals and removing motion artifacts

Worked with multiple groups and dealt with rapid upgrades in the signal detection

circuit with a fast turnaround time

Research Scientist, Prognostics Center of Excellence, NASA ARC (MCT, Inc.), Jan. 2008 –

April 2012

Fault detection, diagnosis and failure prediction for aerospace applications

Explored a wide variety of machine learning techniques including Bayesian learning and

Gaussian Process Regression for battery health monitoring using distributed sensor

SANKALITA SAHA Page 2

networks. The health monitoring system was capable of predicting failure/end-of-life for a

nominally aging battery using intelligent online updating of current state-of-health model.

The implemented system achieved prediction accuracy of above 95% when predicted at

time instances below 80% of total device life

Fault modeling, detection, diagnosis and end-of-life prediction for power electronic devices

under accelerated aging conditions. The designed system used Bayesian learning that

intelligently upgraded the initial physics based failure prediction model online with potential

failure prediction accuracy of above 90% when predicted at 70% of total device life

Feature selection and analysis for anomaly/fault detection and diagnosis and failure

prediction for Li-ion batteries. Analysis of model characteristics and uncertainty

management for Bayesian learning methods applied to batteries and power electronics

Graduate Research Assistant, Dept. of ECE, University of Maryland, 2002 –2007

Design and synthesis for embedded computer vision systems

Designed a new tool for rapid performance estimation and design space exploration of

complex computer vision applications on embedded SoC platforms. Investigated

applications such as face detection and facial pose recognition and tracking in video

Designed and implemented a parameterized design framework for embedded particle filter

based state estimation system was implemented. The targeted applications include particle

filter based facial pose tracking in video

Explored multiple platforms including FPGA multimedia boards (Xilinx ML310), DSPs

(Texas Instrument TMS320c64xx) and shared memory multiprocessor (Sunfire 6800

containing 24 SUN UltraSparc III machines)

Designed and implemented an embedded gesture recognition system based on Hidden

Markov Models with Xilinx Virtex II multimedia board

Technical Intern, Texas Instruments, (June-Dec 2006)

Main designer and developer of project to devise software synthesis techniques for different VoIP algorithms –

Created a framework for efficient software packaging for a variety of DSP platforms for

applications developed in C and DSP Assembly using RTSC (Real-Time Software

Components) packaging tool.

EDUCATION

PhD. (Major: Computer Engineering, Minor: Microelectronics), Dept. of ECE, Univ. of

Maryland, College Park, MD, USA, Fall 2007.

Bachelor of Technology (Hons). (Electronics and Electrical Communication Engineering),

Indian Institute of Technology, Kharagpur, India, Spring 2002.

TECHNICAL EXPERTISE

Programming: C MATLAB R Java OpenMP MPI Verilog DSP Assembly HTML

JavaScript Python

Platforms/Tools: MATLAB Simulink Netbeans Eclipse PyCharm Xilinx ISE

Operating Systems: Windows Mac Linux UNIX

SANKALITA SAHA Page 3

HONORS AND AWARDS

Best paper award at Annual Conference of the Prognostics and Health Management Society

2011. “A Model-based Prognostics Methodology for Electrolytic Capacitors Based on Electrical

Overstress Accelerated Aging”, J. R. Celaya, C. Kulkarni, G. Biswas, S. Saha, K. Goebel

Best paper award at Annual Conference of the Prognostics and Health Management Society

2011. “Investigating the Effect of Damage Progression Model Choice on Prognostics

Performance”, M. Daigle, I. Roychoudhury, S. Narasimhan, S. Saha, B. Saha, K. Goebel

Best IEEE GOLD paper award at IEEE International Conference on Prognostics and Health

Management (PHM) ‘08. “Metrics for Evaluating Performance of Prognostic Techniques”, A.

Saxena, J. Celaya, E. Balaban, K. Goebel, B. Saha, S. Saha and M. Schwabacher

SELECTED PUBLICATIONS

Book Chapters

K. Goebel, A. Saxena, S. Saha, B. Saha and J. Celaya. “Prognostic Performance Metrics”. Book

Chapter, In A. N. Srivastava and J. Han, editors, Machine Learning and Knowledge Discovery

for Engineering Systems Health Management, CRC Press, 2011

S. Saha, and S. S. Bhattacharyya. “Design methodology for embedded computer vision

systems”. Book Chapter, In B. Kisacanin, S. S. Bhattacharyya, and S. Chai, editors, Embedded

Computer Vision. Springer, 2008

Journals

S. Saha, N. K. Bambha and S. S. Bhattacharyya, “Design and Implementation of Embedded

Computer Vision Systems based on Particle Filters”, Elsevier Journal on Computer Vision and

Image Understanding, 2010

S. Saha, V. Kianzad, J.Schlessman, G.Aggarwal, S. S. Bhattacharyya, W.Wolf, and R.Chellappa,

“An architectural level design methodology for smart camera applications,” International

Journal of Embedded Systems, 4(1):83-97, 2009

S. Saha, S. Purayil, J. Schlessman, S. S. Bhattacharyya, and W. Wolf. “The Signal Passing

Interface and its Application to Embedded Implementation of Smart Camera Applications”. In

Proceedings of IEEE, Special Issue on Distributed Smart Cameras, 2008

Conferences

A. Saxena, J. R. Celaya, I. Roychoudhury, B. Saha, S. Saha, K. Goebel “Designing Data -Driven

Battery Prognostic Approaches for Variable Loading Profiles: Some Lessons Learned,” In Proc.

of First European Conference of the Prognostics and Health Management Society 2012

S. Saha, J. R. Celaya, V. Vashchenko, S. Mahiuddin and K. F. Goebel “Accelerated Aging with

Electrical Overstress and Prognostics for Power MOSFETs,” In Proc. of IEEE EnergyTech

2011

S. Saha, B. Saha, A. Saxena and K. Goebel, “Distributed Prognostic Health Management with

Gaussian Process Regression,” In Proc. of of IEEE Aerospace Conference, 2010, Big Sky, MT.

S. Saha, B. Saha and K. Goebel. “Communication Optimizations for a Wireless Distributed

Prognostic Framework”. In Proc. of IEEE Aerospace Conference, 2009, Big Sky, MT

S. Saha, N. Bambha, and S. S. Bhattacharyya. “A parameterized design framework for hardware

implementation of particle filters”. In Proc. of the International Conference on Acoustics,

Speech, and Signal Processing, pages 1449-1452, Las Vegas, Nevada, March 2008

S. Saha, J. Schlessman, S. Puthenpurayil, S. S. Bhattacharyya and W. Wolf, “An optimized

message passing framework for parallel implementation of signal processing applications”, In

Proc. of the Design, Automation and Test in Europe Conference and Exhibition, pages 1220-

1225, Munich, Germany, March 2008

SANKALITA SAHA Page 4

S. Saha, N. K. Bambha and S. S. Bhattacharyya. “Multiprocessor Implementation of a Face

Detection System”, Annual Workshop on High Performance Embedded Computing, 2007

S. Saha, C. Shen, C. Hsu, A. Veeraraghavan, A. Sussman, and S. S. Bhattacharyya. “Model -based

OpenMP implementation of a 3D facial pose tracking system”, in Proc. of the Workshop on

Parallel and Distributed Multimedia, pp 66-73, August 2006

S. Saha, V. Kianzad, J. Schlessman, G. Aggarwal, S. S. Bhattacharyya, W. Wolf, and R.

Chellappa, “An architectural level design methodology for embedded face detection”, In Proc.

of the International Conference on Hardware/Software Codesign and System Synthesis, pages

136-141, Jersey City, New Jersey, September 2005

J. Schlessman, S. Saha, W. Wolf, and S. S. Bhattacharyya. “An extended motion-estimation

architecture applied to shape recognition”. In Proc. of International Conference on Multimedia

and Expo, Amsterdam, Netherlands, 2005

IMMIGRATION STATUS

US permanent resident

Contact this candidate