Instructor @ University at Buffalo IE 374: System Modeling and Optimization: OR IIAn undergrad level course covering Markov Chains, Markov Decision Processes, Queueing Theory, Decision Analysis and Game Theory. From January 2015 to May 2015 (5 months) Department of Industrial and Systems Engineering, Buffalo/Niagara, New York AreaSenior Quantitative Analyst @ Kayson Company • Mathematically Modeled KC Investment...
Instructor @ University at Buffalo IE 374: System Modeling and Optimization: OR IIAn undergrad level course covering Markov Chains, Markov Decision Processes, Queueing Theory, Decision Analysis and Game Theory. From January 2015 to May 2015 (5 months) Department of Industrial and Systems Engineering, Buffalo/Niagara, New York AreaSenior Quantitative Analyst @ Kayson Company • Mathematically Modeled KC Investment Markets (e.g. Iran Housing Market, Energy Market, etc.) using econometric models such as ARIMA, GARCH, Vector AutoregRessive (VAR).• Analyzed financial time series (such as Currency Fluctuations) using NARX Neural Networks and Singular Spectrum Analysis.• Optimized KC Currency Portfolio using constrained nonlinear optimization techniques.• Provided and Evaluated 10 feasibility studies of KC investment projects (i.e. Kahnooj Power Plant, Shirvan GTL Petrochemical Plant, Tehran Shopping Center, etc.-team work)• Simulated KC EPC, EPCF, BOT, Buyback Contracts (i.e. Tehran-Chaloos Highway, a $ 2.5 billion BOT contract, Gole-Gohar Water Transition, $400 million contract-team work). From March 2010 to August 2012 (2 years 6 months) Tehran-IranResearch Assistant @ Computational Intelligence and Large Scale Systems Lab, Tehran Polytechnic • Implemented signal processing techniques to model economical time series.• Analysed Fuzzy time series and compared with Neural Networks.• Studied chaotic systems to improve optimization algorithms (team work).• Applied most popular bio-inspired, evolutionary and meta-heuristic algorithms (team work).• Analyzed performance of clustering algorithms such as k-means, fuzzy c-means, Self-Organized Maps, Neural Networks, Genetic Algorithms and proposed new clustering Algorithm called Noisy K-means (NKM). From 2006 to 2010 (4 years) Tehran, IranData Mining Expert @ Research & Development Center of Bank Mellat • Implemented a Fuzzy Inference System (FIS) to classify bank customers (team work), 2007.• Identified key customer features using Data Mining of customer accounts (team work), 2006. From 2006 to 2008 (2 years) Tehran, IranResearch Assistant @ The Research Foundation for SUNY • Designed and implemented an Intelligent Tutoring System (ITS)¹ that behaves as a self-optimizing tutor and learns the current state of knowledge of students and proposes activities to promote the students’ knowledge (tested in 3 courses). • Developed a Parallel Sparse Matrix Factorization algorithm for Recommendation Systems (Netflix Data) using Coordinate Descent and Stochastic Gradient Descent algorithms (54 million variables, 5x speedup and 60% computational efficiency).• Implemented MapReduce/Hadoop Approach to calculate pairwise similarity between 4 million newborn babies using 2012 U.S. Birth Dataset and construct a graph in order to employ Graph Analytic techniques for community detection, etc. • Implemented graph analytic methods for extracting the social network from a Cumulative Data Graph and identifying High Value Individuals (HVIs) on network-based Hard/Soft data and information fusion (80% accuracy in detecting HVIs).• Proposed a new mathematical framework to model Team Formation Problem using social structures among individuals and developed a heuristic algorithm based on variable depth neighbourhood search (used for networks up to 1000 nodes).• Extracted Latin Authors’ Network from Intertextual Data and clustered the authors based on the network structure. • Employed data mining techniques to discover fundamental, structural and behavioral Laws of Large Social Networks (National Science Foundation-team work).• Developed a probabilistic framework for distilling peer pressure on group decision making.• Reviewed literature and investigated applications of Probabilistic Graphical Models in large social networks.• Extracted and statically analyzed the nurses' social network of a US healthcare center to propose a prescriptive model for the team formation based on social structures. From August 2012 to May 2016 (3 years 10 months) Buffalo/Niagara, New York AreaStaff Machine Learning Scientist @ Twilio Inc. I enjoy building new products using machine learning and optimization techniques for production environment. During past few years, I have been involved in several exciting projects at Twilio including:• Implemented off-line and on-line learning paradigms using LSTM, DeepAR, IForset, filtering methods and time series models to build a real-time anomaly detection system at scale for main calling routes to detect incidents in Twilio’s supernetwork.• Designed and built a scalable ML pipeline to consume more than 50m calls per day from Kafka, and train models (i.e. XGBoost, Gradient Boosting, Bagging, etc.) and score quality of calls in real-time. • Built an end-to-end machine learning-based system (i.e., API) to score quality of conversations in contact centers using advances in speech analytics, natural language processing and deep learning (in tensorflow).• Developed a speech emotion recognition system based on Bi-directional Long Short Term Memory (BiDir-LSTM) to detect angry/unhappy customers (in tensorflow).• Designed and implemented models for text sentiment analysis, text classification and hate/offensive speech detection using Conventional Neural Networks (CNNs) (in tensorflow). San Francisco Bay Area
University at Buffalo
January 2015 to May 2015
Department of Industrial and Systems Engineering, Buffalo/Niagara, New York Area
Senior Quantitative Analyst
March 2010 to August 2012
Computational Intelligence and Large Scale Systems Lab, Tehran Polytechnic
2006 to 2010
Research & Development Center of Bank Mellat
Data Mining Expert
2006 to 2008
The Research Foundation for SUNY
August 2012 to May 2016
Buffalo/Niagara, New York Area
Staff Machine Learning Scientist
San Francisco Bay Area
University at Buffalo
Master’s Degree, Computer Science and Engineering, 4.00/4.00
2015 to 2016
University of Tehran
Master of Science (M.Sc.), Operations Research, 3.9/4.00
2004 to 2007
University at Buffalo
Doctor of Philosophy (Ph.D.), Operations Research, 3.95/4.00
2012 to 2017
IE 374: System Modeling and Optimization: OR IIAn undergrad level course covering Markov Chains, Markov Decision Processes, Queueing Theory, Decision Analysis and Game Theory. IE 374: System Modeling and Optimization: OR IIAn undergrad level course covering Markov Chains, Markov Decision Processes, Queueing Theory, Decision Analysis and Game Theory.
What company does Alireza Farasat work for?
Alireza Farasat works for University at Buffalo
What is Alireza Farasat's role at University at Buffalo?
Alireza Farasat is Instructor
What industry does Alireza Farasat work in?
Alireza Farasat works in the Information Technology and Services industry.
Enjoy unlimited access and discover candidates outside of LinkedIn
One billion email addresses and counting
Everything you need to engage with more prospects.
ContactOut is used by
76% of Fortune 500 companies