SUMMARY:

I am a software engineer with many years of experience in the design and implementation of software and firmware systems. I have worked in a wide array of disciplines/domains and can rapidly and learn new technologies and techniques. I have designed and implemented a both small and enterprise scale systems. I enjoy R&D projects as well as multi - participant collaborative efforts.

SKILLS:

• C
• C++
• Java (core including threads, RMI)
• Python versions 2.X and 3.X
• Objective C
• Mac OSX
• SQL - Analysis, design, application programming, normalization
• BASH Shell Scripting
• UNIX, LINUX
• CMD Batch Scripting
• Windows
• SCCS, RCS, CVS, Subversion, GIT
• UNIX
• Solaris, Ultrix, AIX
• LINUX
• SUSE, Ubuntu, Red Hat, Centos, Fedora
• MAC OSX (UNIX)
• All releases
• IOS (iPhone/iPad)
• Core + GUI
• Windows
• Core + MFC in Visual Studio + C/C++
• Network and Distributed Computing
• Sockets (TCP/UDP/Multicast)
• UNIX, LINUX, Windows, Frameworks
• VOIP (SIP/SDP)
• SIP phone IOS application
• RPC (DCE/ONC)
• Apple CF* core foundation
• Apple NS* network components
• Like NSURL and the like
• Java RMI
• CORBA
• On Solaris
• DCOM/ATL (Microsoft)
• Windows

SPECIAL/ADDITIONAL SKILLS:

• Scientific and numerical computing
• Matlab, Maple, Python (numpy, scipy, scikit-learn), FORTRAN, LINPACK, basic Numerical Analysis
• Multithreaded application design and implementation
• POSIX threads, Java Threads, Python threads and processes, Win32 threads, + common synchronization primitives
• SQL API programming
• ODBC, JDBC, pyodbc (Python), Embedded SQL (Pro*C), SQLITE, C-level dblib-type C libraries
• Microchip PIC-24, PIC-32, dsPIC Firmware design/implementation
• Using MPLAB (MPLAB-X) - Personal hobby work
• Board-level communications
• SPI, Serial, I2C and USB
• Digital Signal Processing
• FFT, STFT, Cepstral analysis, FIR/IIR filter design(with tools), LPC, Audacity and other similar 3 rd party tools
• Computational Linguistics
• HTK(acoustic and linguistic) NLTK(Python)
• Machine Learning
• GMM, K-Means, Nearest Neighbor clustering plus HMM recognition techniques various Matlab tools and Python tools
• Development Environments
• XCode, Eclipse, Pydev, vi-make, iPython,
• Visual Studio (version 8 and greater)
• Object Modeling with UML
• Visio and 3 rd -party graphics tools
• J2EE Glassfish (Java App. Server) Deployment
• NASA WorldWind GIS System Programming and Design for live map annotation.
• Java FX, JSP and Java Servlets

KEYWORD SUMMARY:

C, C++, Java, Python, OSX, MacOS, Objective-C, DSP, Digital Signal Processing, DSP, UNIX, LINUX, pthread, multithread, POSIX, VOIP, FFT, audio analysis, speech recognition, numerical analysis, statistics, SQL, Oracle, MySQL, Postgresql, SQLITE, Matlab, Maple, computational linguistics, database programming, database schema design

PROFESSIONAL EXPERIENCE:

Software Engineer

Confidential, St. Petersburg, FL

Responsibilities:

• Design of firmware for the control of disinfection cycle of the hemodialysis machine involving a variety of pumps, valves and sensors to be monitored in near real-time.
• C/C++ as the language of implementation language on both the INTEL development environment and the targeted TI microcontroller sides of development.
• Implementation environment is TI’s Code Composer Studio, which is Eclipse-based.
• I am also responsible for unit testing a variety of development artifacts using the CPPUnit testing framework.
• Core system does not have an operating system thus requiring design and coding techniques suitable for resource-constrained systems.

Software Engineer/Data Scientist

Confidential, St. Petersburg, FL

Responsibilities:

• Principal tools: Python (NumPy, SciPy, Scikit-Learn), C/C++ (SVM, frequency-domain signal analysis, deployment)
• Creating speaker identification and speaker diarization models in support of comprehensive analysis of voice in phone communications.
• Employment of machine learning techniques in creation of classifier models (Multi-class Support Vector Machines, Gaussian Mixture clustering, Kernel Ridge Regression, etc.)
• Conducting research and created several working models for the detection/inference of speaker emotional state from recorded audio.
• Developing audio capture software to pull call and audio data from VOIP/SIP telephony systems for audio analysis real-time.
• Operating environments: Windows, LINUX, MacOS

Senior Software Engineer

Confidential, Tampa, FL

Responsibilities:

• Principal work involved programmatic creation and manipulation of geometric objects for the NASA WorldWind GIS system for globe/map annotation. Language of implementation was Java 8 and involved the implementation of several geometrical transforms and algorithms. Principal users of the objects are military and D.O.D. personnel.
• Work also involved JAXB marshaling/unmarshalling of Java geometric objects to back-end systems.

Senior Software Designer

Confidential, Clearwater, FL

Responsibilities:

• Proscript is a medical dictation/transcription service provider for hospitals and medical clinics.
• My principal task was to support the speech recognition back-end which supports the creation of transcripts from dictation audio. As such, my tasks encompassed language modeling (computational linguistics) as well as acoustic modeling (frequency domain signal features).
• Remapped the language model in-memory image to optimize the loading of language models which are often > 9 GB in size.
• Employed compression algorithms to minimize the network demand for moving large language models across NFS shares.
• Using Python, and the open source package NLTK, I constructed and trained a POS tagger and a document component recognizer as part of the speech recognition audio to text process.
• Using Python with the open source packages (numpy, scipy) I constructed and trained a speaker classification model using Gaussian Mixture Models clustering (GMM) and subclassing to better match up speakers who are acoustically similar. This aids in creating adaptive acoustic models.
• Derived some useful and novel acoustic features for modeling using Short-Time analysis techniques including pitch tracking and Linear Predictive Coding( AR LPC) for silence, voiced, unvoiced speech and formant estimation as prime speaker class discriminants.
• Adapted the Kaldi speech processing system to operate within Proscript’s operating model.
• Designed an acoustic analysis module using Python and open source packages to explore dynamic speaker classification and dynamic adaptation as part of the speech recognition process (preprocessing of incoming text-independent speech)

Senior Software Design Engineer

Confidential, Clearwater, FL

Responsibilities:

• Duties included design and implementation of software and firmware for Confidential fueling solutions
• Primary platform was an embedded LINUX system
• Primary focus was on machine-machine communications and data stream integrity and message throughput optimization
• Development was in C (ARM) cross compiled in the Windows environment under Cygwin with Eclipse and GDB remote debugging

Senior Programmer

Confidential, Sarasota, FL

Responsibilities:

• The task Confidential hand was the development of a VOIP phone application (like Skype) on the iPhone using IOS
• XMPP was used for contact/presence management
• Significant hurdles imposed by Apple’s background processing model were overcome
• Principal languages were Objective-C and C/C++ in the XCode programming environment.
• Apple push notifications were implemented efficiently
• Real-time network performance and call quality monitoring comprised a significant percentage of the development effort

Contract Programmer

Confidential, Tampa, FL

Responsibilities:

• Principal task was the creation of a new network connectivity/security application on Mac OSX to enable ease of communication setup by end users.
• The application architecture was event-driven.
• IPSec VPN tunnel was used for secure communications and involved integration with Apple’s core VPN and networking services.
• I also designed and implemented the user interface (GUI)
• An interface to cellular modems was also designed and implemented.
• Principal tools were Objective-C, Apple Foundation classes and third-party libraries
• The implementation of the WISPr 1.X Wi-Fi secure interface was a significant challenge and accomplishment.

Contract Programmer

Confidential, Oldsmar, FL

Responsibilities:

• Designed and implemented the conversion of a large, fork-parallel file transfer server to a multithreaded server in order to optimize scheduling and resources on the Solaris platform. The conversion to a multithreaded architecture allowed the server to scale dynamically and adaptively depending on the offered process load.

Scientific Programmer

Confidential, Dunedin, FL

Responsibilities:

• Designed and implemented a small interpreter and language suitable for programming a small, portable spectrometer.
• The interpreter was written in C using YACC and LEX compiler generation tools and the language exposed all features of the device in a consistent manner.
• The interpreter could be run from MAC OSX or LINUX or Windows and delivered its byte-code to the handheld spectrometer.

Contract Programmer

Confidential, St. Petersburg, FL

Responsibilities:

• Very short-term contract for small code enhancements and unit testing of same for airport X-ray package scanners.
• Languages were C and C++ with custom libraries linked in for a LINUX-based target.