Hugo Louro's Email Addresses & Phone Numbers

📖 Summary

Apache Storm PMC, Committer @ The Apache Software Foundation Contributor to Streaming and Big Data Apache projects (Kafka, Storm, Flink, HBase, Ambari, Ratis). Active in the community leading initiatives, voting for new committers, reviewing code, testing releases, and presenting at Big Data conferences and Meetups. Streaming Data Architect @ RealFevr Driving streaming technology adoption; evaluating Kafka Streams, Flink, Beam, Spark Streaming. Contributing to Flink. Prototyping streaming pipelines to generate searchable news for the web/phone app, and to compute statistics, analytical insights, and predict players’ fantasy scores for semi-structured, unbounded, and out-of-order player and game data.Advocating for the redesign of the data ingestion workflow to stage live match data feeds in Apache Kafka to improve scalability, fault tolerance, decoupling, and eliminate data loss. Participating in the next generation microservices architecture design discussions and in the evaluation of Docker and Kubernetes to simplify cloud deployment and resource management. Staff Software Engineer @ Hortonworks Initial member of the Streaming Platform team. Co-creator of open source technologies. Contributed to Streaming and Big Data Apache projects. Participated in company-wide software architecture reviews and Cloud/infrastructure strategy discussions as member of the Architects’ group. Lead internships and mentored engineersR&DStormRecreated Kafka sources to enable distributed, fault tolerant, ingestion of streaming and batching data with improved throughput (+70%) and reduced latency (-30%). Implemented a machine learning processor (PMML Bolt), and a Bolt/Trident State to support reliable streaming/batching writes of documents into Apache SolrKafkaProposed backwards compatibility improvements to the “new“ Kafka consumer client API. Coded producers/consumers in Java. Contributed to releases with code reviews, testingStreaming Analytics Manager / Schema Registry Co-creator, co-architect, and first beta release co-manager; designed features, wrote code, and worked with PM, QE, RE, and tech writers. Developed clients/REST APIs to expose metadata for Storm, Kafka, Hive, H-Base, HDFS, Zookeeper in secure/insecure clusters (Kerberos, SPNEGO, SSL). Created a distributed, horizontally scalable, Rules Engine that enables users to define complex rules at design time and executes (streaming SQL, Groovy) with pluggable runners (Storm, AthenaX/Flink, Piper/Airflow). Developed ORM framework to persist entities into pluggable databases (MySQL, Postgres, Phoenix, H2) to unify data persistence across relational, row-oriented, and non-relational, column-oriented, SQL/NoSQL databases. Built cache Facade to facilitate memoization of pipeline results into pluggable caches (Guava, Redis, Hazelcast)Engineering AssistanceProvided assistance 24/7 during support escalations of large Big Data clusters in issues such as security (Kerberos, SSL), Thrift packets integrity, network breaches, Kafka MirrorMaker performance, under replicated partitions, unclean leader election, quotas, ACL From June 2015 to March 2018 (2 years 10 months) Senior Software Engineer @ AppDynamics Early engineer in the Application Intelligence Platform; contributed to create a new platform, maintain legacy code, and helped expand the team to 100+ engineers during hypergrowth phase (200-1000 employees) in anticipation of IPOR&DContributed to create a distributed, horizontally scalable, zero downtime services-oriented platform to replace the legacy, monolithic, concurrent, persistent, on-premises and multi-tenant cloud backend. Participated in several aspects of the definition, design, implementation, and technology stack choices for the Metadata (Hazelcast), Metric Data (HBase), and Event Processing (Elastic Search) services. Created Java and REST APIs that support versioning, media types, and multiple serialization protocols.Contributed several features and enhancements to the legacy platform (Glassfish, J2EE, MySQL, Hibernate, EhCache). Namely, contributed to optimize MySQL query reads, to redesign the alerting system to evaluate rules concurrently, to solve racing conditions and cache inconsistencies, to support email templates, and to decouple data and business layersEngineering AssistanceBuilt a thread-safe logging extension (Slf4J) that throttles and filters log events for target entities specified interactively; facilitates debugging in production and reduces support response time during escalations on-premises and in Cloud. Solved multiple scalability and performance bugs in SAAS by resorting to Splunk log analysis, SSH login onto remote servers/VMs, and collection and analysis of thread and heap dumps From September 2013 to February 2015 (1 year 6 months) San Francisco, CASenior Software Engineer @ TIBCO Software Inc. Contributed to Complex Event Stream Processing and Rules Engine technology TIBCO BusinessEvents (BE) and to distributed data grids as member of the Quantum Leap R&D team.R&DCo-created BE’s Monitoring and Management (MM) server (Java, RMI, JMX, Flash) to provide metrics, alerts, and administration capabilities to distributed data grid clusters of Oracle Coherence and TIBCO ActiveSpaces. Additionally, developed a pluggable, JMX compliant, JAAS security module that supports SSH, password and private/public key authentication, and file-based roles authorization. Instrumented query streams and ontology objects with JMX MBeans (Standard, Model).Co-developed a connector to TIBCO Spotfire TERR (predictive analytics platform) that delivers low latency, high throughput, and high parallelism analytics to streaming (business) events.Acquired hands-on experience with distributed data grid cluster provisioning, messaging systems (JMS, TibcoRV), orchestration, and business process frameworks (BW).Developed an OPC protocol driver client in C# that enables reliable two-way TCP communication between BE and the Smart Grid; component deployed as part of the Xcel Energy SmartGridCity project. From February 2008 to July 2013 (5 years 6 months) Palo Alto, CAVisiting Researcher @ DAGS Group @ Stanford University Researched Object Detection in cluttered scenes with Support Vector Machines trained with a mutual context model (action, pose, context information) for higher classification accuracy; latent variable parameters learned with self-paced learning From 2010 to 2011 (1 year) Visiting Scholar @ Stanford Mood and Anxiety Disorders Laboratory @ Stanford University Studied classification algorithms featuring volumetric and morphometric fMRI data to categorize prevalence of MDD. Coded Python and SPSS scripts for statistical analysis of fMRI data. Contributed to imaging software (DtiQuery, MrDiffusion, MrVista) From 2008 to 2010 (2 years) Research Associate @ Translational and Developmental Neuroscience Laboratory @ University of Michigan Researcher, single developer of the first code implementation of Context Dependent Correlation Analysis (PPI) of fMRI data. Project in collaboration with the NIH that resulted in a highly cited journal publication in PPI [4]. Developed Python, Shell, MATLAB, SPM, AFNI scripts to analyze fMRI images and convert medical imaging formats From 2006 to 2008 (2 years) Research Fellow @ Communications and Signal Processing Laboratory @ University of Michigan Researched the application of relevance feedback methods to Bayesian Networks to enable expert assisted discovery of gene regulatory networks. Applied Kernel estimators with ad-hoc and automatic procedures to estimate probabilistic models of univariate and multivariate data sets of genomic data without assuming prior knowledge of the distribution. From 2004 to 2006 (2 years)