Abraham Aldaco-Gastelum
San Ramon, CA
aczezy@r.postjobfree.com
SUMMARY
Over 8+ Years of experience in Applied statistics, Design of Experiments, simulation, modeling and developing custom algorithms for statistical Analytics and developing applications with statistical and mathematics libraries. Over 2+ Years developing applications with Machine Learning supervised algorithms and libraries. And over 15+ Years of experience in IT Industry - administration, architecture, designing, configuration, testing, troubleshooting, and supporting LAN/WLAN/WAN networking technologies, with in-depth knowledge of network architecture and protocols.
TECHNICAL SKILLS
Algorithms:
Design of Experiments (DOE), Full, Fractional, Saturated, Optimal, Covering arrays
Machine Learning (ML), supervised classification, regression, k-NN, Naive Bayes, Decision trees algorithms
Markov Chain, the two states Gilbert–Elliot model, State Transition Diagrams
Linear regression multivariate, Ordinary Least Squares (OLS), Weighted Least Squares (WLS)
Variable selection Forward (FS), Backward elimination (BE), Stepwise
Predictive Modelling
Develop custom algorithms for finding, among numerous variables, the most significant variables and interactions
Test algorithms robustness
Software:
Strong Computer Science fundamentals (algorithms and data structures)
JMP (SAS), SPSS, Design Expert, Minitab, R, MS Excell, Octave, Latex, Gnuplot
C/C++, Java, Python, TCL/TK, AWK, Linux Shell, MySQL
Python numpy, scipy, pandas data analysis libraries, matplotlib plotting
Workloads for characterization and benchmarking (Filebench, Netperf, TCPDump)
Script simulations, Mining BIG data traces, select variables, construct predictive models
Discrete SIMULATOR NS-2, NS-3
Oracle VM VirtualBox, Vmware virtualization for Desktop and Server
Operating Systems: Linux, MS Windows
Statistics:
Statistics t-Test, z-Test, u-Test, F-test, hypothesis testing, p-Value, ANOVA, Correlation, Multicollinearity, Aliasing determination, Eigenvalues
Numerical (Discrete, Continuous), Categorical (ordinal, nominal, interval) factors
Model selection Akaike Information Criterion (AIC) and the Bayes Information Criterion (BIC)
Parametric, non-Parametric data distribution
Skewness (lack of symmetry) and kurtosis (tails, or outliers)
Shapiro-Wilk, Kolmogorov Smirnov tests to determine the normality of frequencies
Non-parametric Man-Whitney U-test, Wilcoxon Sum Rank Test
Matrix computations Correlation (X X), Variance Inflation Factor (VIF) (X X) 1
Coefficient of determiation R-squared, Adjusted R-Squared, Mean Squared Error (MSE)
Transformation for observed data, Shapiro-Wilk normality verification
Plots normal probability plot, measured vs fitted, residuals vs fitted, histogram, Box-plot, 1/2
Cisco Certified Network Associate (CCNA), in-depth knowledge of Network protocols
Cybersecurity
Agile methodology, Git, Dimensions, Rally, Doors, Jenkins, Scrum
Strong research capabilities
PROFESSIONAL EXPERIENCE
IBM, Guadalajara MX October '2015 – December ‘2016 Data Test Storage Performance
Responsibilities:
Use and deep understanding of Machine Learning k-NN, Naive Bayes, Decision trees algorithms
Use and deep knowledge in Statistics and Design of Experiments (DoE)
Strong knowledge of predictive Linear regression modeling
Data extraction and Cleansing
Test the performance of I/O operation in datacenter servers of high storage capacity (virtual storage IBM cache TS7720T, IBM TS3500 Tape Library and LTO tape drive)
Use best transformation of data to improve normality distribution of the performance
Benchmark the performance
Develop test plan based in Experimental Designs
Find the most significant Factors for the performance
Compute coefficient of determination R-squared to measure correlation between Factors and the response, the performance
Write a white paper of the new release showing the Analytics from the results: include Plots and quantitative description of new improvements
Linear regression modeling of the performance
Show Statistical Analytics from hardware improvements compariring with previous versions
Execute Functional and Systems validation
Find and report defects from software and hardware and expose defects and problems in phone meetings
Publish defects and updates in a triage system for reviewing by technical experts
SCRUM methodology
General Electric, Queretaro MX April ‘2015 – October ‘2015 Project Engineering Leader Test Aviation Systems
Responsibilities:
Test performance from Write/Read operations on flash memory
Improve quality of software (C) by finding and reporting defects during testing and code reviewing
Construct Analytics from the output, the performance
Show the analysis in plots, tables and text
Search for abnormal patterns from the analytics and report
Plot delay in writing 512 bytes size sector flash memory, including standard deviation from the mean (repeat for reading)
Develop software for embedded devices for avionics
Update user requests documentation (Doors)
Lean/Six Sigma Certification (in progress)
AGILE, Scrum methodologies (Rally, Jenkins) and version control software (Git and Dimensions) Arizona State University, Tempe Arizona August ‘2008 – April ‘2015 Data scientist in complex systems with numerous factors Responsibilities:
Strong experience in Researching
2/2
Develop an innovative algorithm for screening Experimental Designs in complex systems with numerous variables, use of parametric and non-Parametric statistic tests
In-depth knowledge of Design of Experiments (DOE)
Construct or select most appropriate DOE considering Size, coverage of ALL main effects and low-order interactions
Algorithm allows:
Study numerous factors while traditional approaches are ineffective
Use hypothesis testing
Include multiple levels per factor
Have big/large experimental design
Find the most significant factors and factor interactions first
Obtain the higher coefficient of determination ( R-squared )with the least number of factors
Construct linear regression models of high predictive capability
Construct predictive models including only the most significant variables, compute R-squared, AIC, BIC as stopping criteria
Design a custom methodology for variable selection based on the variability of factors and interactions to determine significance.
The methodology comprises:
Analysis of the best transformation of the observed data
Grouping by similar variability
Deep analysis of the difference on variance in data from factors frequency
Analysis of the data distribution (e.g. Normal distribution) and indepence of the factors
Man-Whitney U-test and Wilcoxon Sum Rank Test for Non-parametric
Difference of means t-Test for Parametric
Ordinary least squares (OLS) or Weigthed least squares (WLS) as required
Automate large simulations and perform analytics from the BIG data traces
In-depth knowledge of discrete SIMULATOR NS-2, familiar to NS-3
Modify simulator source code to customize functioning
Develop scripts (TCL/TK) to automate simulations, going from minutes up to days
Bind the TCL variables to NS-2 variables to control different value per variable per each run
Develop scripts (AWK) to compute performance (bits/seconds) from numerous BigDATA traces
Write and publish scientific Journal papers
Construct a case study of a complex systems in mobile ad-hoc network (MANET)
The case study includes:
A wireless mobile ad-hoc network (MANET) using NS-2 discrete simulator
75 controllable Factors across-layers from Network Architecture of protocol stack, Error, Energy and Propagation and Random Waypoint Mobility models were included
Machine Learning techniques to simulate the Error model for packet transmission and reception (The Gilbert–Elliot model, a Markov Chain model of two states)
Covers at least all 2-factor interactions among the 75 factors
Construct a TCP throughput prediction model
Linear regression model with 10 terms containing only 9 most significant factors out of the 75
Includes main effects and 2-factor interactions (R-squared = 87%) Arizona State University, Tempe Arizona June ‘2013 – June ‘2014 Data analyst at Department of Speech and Hearing Science Responsibilities:
Managed a statistical data of a 5 years longitudinal study in the Language and Reading Research Consortium
(LARRC) designed to increase our understanding of language- and reading-comprehension development
(1200 children ages 4-8 years old)
3/2
Extensive use of MS Excel, complex formulas, charts, inter Sheet operations, Sort, Vlookup, Hlookup, Extensive formatting to customize the viewing, and statistics Data Analysis tool
Deploy Scan to Database application, design of forms, configure OCR, configure data base connection, supervise scanning process
Define policies to execute systematic cleanning of data in DataBase from the scannig process
Provide analytics of the findings to write scientific journal papers
Construct histogram diagrams to show normality distribution, skewness (lack of symmetry) and kurtosis
(tails, or outliers), and apply Shapiro-Wilk test to determine the normality of frequencies
Extensive use of SPSS Descriptive statistics mean, median, mode, standard deviation, variance, range, skewness, kurtosis, histogram
Administration of Database, videos and electronic tests, classification and MS SharePoint
Show Analytics to Board of directors
Physical custody of confidential documents and tests, classification, sorting, labeling, and electronic processing assurance
Intel Corporation, Chandler Arizona January ‘2012 – May ‘2012 Data analyst and Network performance test
Responsibilities:
Test the performance response of servers using solid state storage devices (SSD) functioning as cache device
Simulation & Emulation
Characterize workloads emulating several clients - one server architecture in Network File System (NFS) in Linux
Characterize workloads using NetPerf and FileBench to test the cache server response and measure the performance in the Network
Capture network traffic using TCPDump
Construct Analytics, plots using Gnuplot, and description of the improvements
Prepare Benchmark of the results for customer presentation Tec de Monterrey Engineering Department, Hermosillo MX May ‘2002 – August ‘2008 Network Engineer
Responsibilities:
Teach courses to undergraduate students and industry staff on CCNA, operating systems, and programming
20% from the total students passed CCNA exam
Administer DNS, DHCP and the IP addressing
Provide support in monitoring 3Com and Cisco switches
Network infrastructure support to routing and switching equipment
Provide support to Back up Cisco IOS to a TFTP, upgrade and restore Cisco IOS from TFTP server
Troubleshoot and resolved Ethernet issues: monitor traffic, detect malicious traffic, patch servers
Troubleshoot ACL issues: denied ports, allowed subnetworks
In-depth knowledge of cybersecurity issues
Use Microsoft Visio to document network design
Tec de Monterrey IT Department, Hermosillo MX January '1997 - May '2002 Network Engineer
Responsibilities:
In charge of a Datacenter including 20+ Linux and Microsoft servers, LAN/WiFi/WAN and 5 remote sites of the University campus in the North West of Mexico
Administration, architecture, engineering and troubleshooting tasks for the core-network and branch locations
Configuration of the WAN including routing protocols, security and E1 WAN links and E1- Ethernet
Build and develop strong relationships with end-users by providing reliable customer support
Configure and monitor servers in a data center and remote sites (linux Red Hat, AIX and MS Windows 2003) 4/2
Configure and administer network services DHCP, FTP, DNS, VPN, NAT, PROXY and MS Active Directory
Strong understanding of ICMP, ARP, UDP, TCP/IP, AODV, IEEE 802.11b
Implement the Routing protocols (RIP, IGRP, EIGRP, OSPF) Cisco routers 2500, 2600 series
Administered network services and systems which monitor computational resources and network traffic
In-depth knowledge of cybersecurity issues
Implement network protocols traffic analizer using Etherpeek, Ethereal, Wireshark, and web traffic FWL 1989 – 1997
Network Engineer and Consulting
Responsibilities:
Design and configure WAN links for data and VoIP communications for central site and 4 remote offices
Develop a inventory and billing control for an automotive retailer
Develop an accounts receivable software for a Departments store EDUCATION & CERTIFICATIONS
Ph.D. in Computer Science, ARIZONA STATE UNIVERSITY, Tempe, AZ, GPA 3.63 2015
M.S. in Computer Science, TEC DE MONTERREY, Mexico, Summa Cum Laude GPA 3.9 2000
B.S. Electronic Systems Engineering, TEC DE MONTERREY, Mexico, GPA 3.5 1988
Cisco Certified Network Associate (CCNA) 2004
Lean/Six Sigma Green Belt Certification (in progress) 2015 5/2