Huang P. personal email
- Valid
Huang P. phone numbers
I help organizations extract maximum value from their data by building efficient data systems using modern data tooling and techniques. I've worked with many companies in the Silicon Valley and bring a lifetime of experience in tech and top-notch engineering skills.Please see my research on the Modern Data Stack:https://medium.com/@huangpan/modern-data-stack-2023-ab3364b9281dAbout me----------Seasoned Polymath Tech Veteran with proven ability to quickly ramp up on new technologies and deliver high quality results. Greatest strengths: entrepreneurial mindset, practical problem solving using modern technologies. Strong track record and passion for building out new projects.Main interests: data architecture & engineering, blockchain / decentralized finance, FinTech, trading- Founding data engineer at Yuga Labs and Roofstock: built up their initial data infrastructure from scratch using the best available technologies at the time- 12+ years of recent software development experience specializing in early stage startups; 10 years of hardware development experience- Expertise in data infrastructure & engineering, financial engineering, statistics / analytics, digital signal processing (time series), telecom / geospatial data- Domain knowledge of cryptocurrencies, FinTech, real estate, lending, credit scores, algorithmic trading systems, quantitative finance- ASIC / FPGA development: 5 years designing 3D GPUs; 5 years designing 4G wireless chips- Graduate of UC Berkeley BS EECS, Stanford MS EE, UCLA Anderson School of Management MFE, Blockchain University; US Citizen
-
Chief Data OfficerPondCalifornia, United States -
Head Of Data EngineeringPond Jul 2024 - PresentNew York, Us -
Principal Data EngineerYuga Labs Nov 2023 - May 2024Yuga Labs is the #1 brand (Bored Ape Yacht Club, CryptoPunks) in the cryptocurrency NFT (Non Fungible Token) space, with $450m in seed funding.- I was the founding data engineer at Yuga and lead data engineering strategy / managed the data infrastructure at Yuga- I was the 2nd member of a data team of 4 consisting of the VP of Data, myself, and 2 data scientists- I built out Yuga's initial data infrastructure from scratch using the best available tools in the Modern Data Stack- With modern data tooling and a small team of senior data professionals, we were able to iterate quickly and achieve a very fast time to business value - key executives were able to make data driven business decisions 1-2 months after project inceptionData tech stack: optimized for speed of development and general scalability with reasonable operating costs- Snowflake data warehouse, Y42 dbt data modeling & data pipeline orchestrator, Sigma Computing BI tool, AWS S3 / Kinesis Data Streams & Firehose, Airbyte, Hex- Blockchain data from Flipside Crypto (Ethereum & Solana), Yuga gaming telemetry (streaming data), social data (Sprinklr)- Annual operating expense of data infrastructure: < $100k / year * Saved Yuga Legal $100k in outside vendor costs by implementing in-house solution using above data infraI joined Yuga when they had ~150 people. Unfortunately Yuga went through another round of layoffs at the end of April 2024: 1/3 of the company was cut, including most of the data team. -
Data Engineering ConsultantFreelance, Self Employment Apr 2023 - Nov 2023No Name, OoProjects:- cambercloud.com (July - Nov 2023) * Worked on bespoke implementations of JupyterHub github.com/CamberCloud-Inc * OO python programming on MVP features; Kubernetes, Docker, Spark 3.3+; AWS S3, EKS- datakind.org (Aug - Nov 2023) * Population demographics from ACS data / census tract shapefiles using Noteable, dask- criticalriver.com (Sept - Oct 2023) * Created MDS reference data architecture for scaleable data extract from Oracle Fusion ERP Cloud to GCP BigQuery * Created modern data pipeline github.com/huang-pan/ny_citibike_pipeline using Astronomer Airflow 2.7+, data modeling & test with dbt 1.6+ / dbt-expectations; GCP Cloud SQL, Cloud Storage, BigQuery; Github Actions CI / CD- greenstand.org uses Airflow on a digitalocean.com K8s cluster and a PostgreSQL DB * Airflow vs Prefect analysis: github.com/huang-pan/treetracker-prefect * Added Airflow DAG development instructions: github.com/Greenstand/treetracker-airflow-dags * Added instructions on how to install Airflow on K8s using Helm and Ansible: github.com/Greenstand/treetracker-infrastructure/tree/master/airflow- Fixed issues github.com/Greenstand/treetracker-airflow-dags/issues * 138: removed duplicate data in Airflow DAG SQL query for BI dashboard * 135: Added Slack notifications to Airflow * 92: Added RBAC to Airflow using K8s secretsUpdated data engineering (DE) skillset (May - July 2023)- Course certificates & notes: https://github.com/huang-pan/modern-data-stack-2023- Completed DE / MLOps courses on: * dbt, Spark 3+, Kafka, FastAPI, Generative AI, K8s, Terraform * AWS: S3, Redshift / Athena, Kinesis, EMR, DynamoDB, Glue, Lambda, Timestream, Sagemaker Studio * Azure / Databricks: Databricks Delta Lake / Delta Live Tables / Unity Catalog; Azure Data Lake / Data Factory / DevOps * GCP: BigQuery, Dataflow, Pub/Sub, Cloud Storage, Cloud Dataproc, Bigtable, Cloud Build, Vertex AI -
CaregivingCareer Break Dec 2020 - Mar 2023Took 2+ years off from tech for personal / family reasonsDec 2020 - Aug 2022:- Elderly care; infant care- Completed real estate projects: completely remodeled / renovated / rented out 4 propertiesSept 2022 - Mar 2023:- Took calculated risk to pursue one of my passions: trading crypto full time. Grew as a trader, but missed the constant learning of tech, so decided to switch back to tech. Crypto trading business relegated to side business with variable income.
-
Senior Data EngineerJuvo Feb 2019 - Nov 2020San Francisco, California, UsJuvo is a global startup that provides airtime lending and Financial iDentity as a Service (FiDaaS). Juvo creates a credit score for the world's underbanked using mobile Telecom data. I developed data products for Juvo and significantly reduced the amount of technical debt in Juvo's data infrastructure. In November 2020 Juvo closed its SF office.Data Products- FiDaaS credit score - Juvo's core data model: wrote Airflow DAGs that create SQL based loan repayment metrics from LATAM telco data. Scores are served through the Juvo Scores API.- FinTech customers (Brazil): Nubank, C6, SuperSIM, etc. * Backtested FiDaaS scores using historical telco and bank loan dataData Ingestion- Voice / SMS / mobile data usage from major Telco partners: Claro Brazil, TIM Brazil, BSNL (India), Tunetalk (Malaysia), etc. * Wrote many Airflow DAGs (python / SQL / PySpark) that ingest above data into Redshift / Snowflake- Juvo airtime lending data (Kinesis to S3 / Redshift)Technical Debt Reduction- Existing data infrastructure: AWS S3 / Redshift / PostgreSQL RDS / Kinesis Data Firehose / DynamoDB, multi-region AWS VPCs (for regional data compliance / privacy), Spark- DevOps infrastructure: Kubernetes / Helm, Docker, Terraform, Ansible, TeamCity CI, Rundeck CD, Datadog- Piecemeal upgrades over time * Introduced Snowflake (replaced Redshift), Sigma Computing (replaced Looker) to Juvo * Rewrote many Airflow DAGs to use python 3 and Snowflake after the LATAM Airflow server was migrated from python 2 to 3 and to AWS EKS- Created a company wide data catalog using AWS Glue- Updated technical documentation, coding guidelines, etc.Created Data Architecture 2.0 roadmap- Old (crashed often): Airflow (python 2), Spark 2.0 / JupyterLab on Juvo managed Kubernetes- New: Airflow (python 3, K8s Executor) on AWS EKS, Spark 3.0 on AWS EMR, AWS SageMaker or Saturn Cloud- Introduce dbt Cloud for SQL data model standardization, Neo4j for cross telco profile aggregation, MLflow for MLOps -
Senior Data EngineerRoofstock Mar 2017 - Jan 2019Oakland, California, UsRoofstock is a FinTech startup disrupting the Single Family Rental investment industry. I lead the efforts to build out Roofstock's core data systems using state of the art tools. I was the founding data engineer at Roofstock and the 2nd member of the data team. The scalable data infrastructure I implemented handled Roofstock's growth from 60 people to 200 and beyond.Data Infrastructure- Set up and maintained Snowflake, Fivetran, Domino Data Lab, Sigma Computing- Spun up open source based Airflow and JupyterLab servers on Azure VMsELT / Reverse ETL- Wrote Airflow DAGs (python 3 / SQL) that: * Ingest terabytes of real estate data from SFTPs, APIs, etc. into Snowflake * Move data from Snowflake to Azure SQL DB using Azure Data Factory; optimized Azure SQL DB tables with 100s of millions of rows to handle fast queries- Structured Snowflake databases; used dbt for data modeling / test in ingestion pipelines- Used Fivetran to ingest data from Mixpanel, Hubspot, Salesforce, etc. into Snowflake- Designed Global User ID integrating disparate data from Mixpanel -> SalesforceData Services- Wrote part of the data services API (.NET Core C#) that serves up US property tax & deed records- Developed several self-service data tools using Domino Launchers- Provided ad-hoc analytics support (SQL queries) to sales, etc.Data Products (Operationalized Data Science)- Data mining large Property Owners from tax / deed data * Sped up the hierarchical clustering algo in the Ownership entity resolution process by adding the python graph-tool library * Wrote several Airflow DAGs that deployed this process to prod; the resulting top owners tables were used by sales to significantly improve Roofstock revenue- Improved the Neighborhood Score algo by adding geospatial data (census tract shapefiles)Miscellaneous- Adopted data sources: real estate (ATTOM, Zillow, etc.), demographic / census, GreatSchools, SpotCrime, etc.- Hired and managed a data engineering intern -
Self StudyData Science And Engineering Jul 2016 - Feb 2017Data Science and Engineering Self Study- Completed a retrospective of everything that could have been done better at Plum Lending- Did a deep dive into modern data engineering technologies and data science algorithms- Interviewed at and completed take home assignments for various companiesOct 2016 - Feb 2017- Exploratory data analysis, predictive modeling, and data architecture assignments for various companies * Using Spark 2.+ / PySpark on Databricks; Python scikit-learn, pandas, numpy, matplotlib, etc. on Jupyter notebooks, Python Flask API for crypto exchange order book * For example, please see: https://github.com/huang-pan/shift for sample data science work using Spark: exploratory data analysis, linear regression, gradient boosted tree regressionJuly - Sept 2016- Ramped up on the latest Big Data technologies: attended many data / machine learning meetups and conferences- Self-study of: * Data Engineering: Snowflake, Airflow, Domino Data Lab, Data Robot, Hadoop, YARN / Mesos, Kafka, ElasticSearch, etc. * Data Science / Deep Learning: regression, classification, clustering, dimensionality reduction, neural networks, etc. * Web Scraping: Requests, Beautiful Soup, Scrapy, Nutch, Selenium * Natural Language Processing: NLTK, dedupe, TextBlob, Named Entity Recognition, word2vec, TF-IDF, topic / sentiment analysis, etc.
-
Vp Of EngineeringPlum Lending Oct 2015 - Jun 2016San Francisco, Ca, UsPlum Lending is a FinTech startup revolutionizing the small balance Commercial Real Estate (CRE) lending space. I was brought on to oversee all technology (both data and web) developed at Plum. While at Plum I started and established the Engineering Department. This was a mostly managerial position.Data Mining / Engineering- Researched & combined CRE data for upload to Salesforce CRM for Inside Sales Department * Data sources: Commercial Mortgage Backed Security data, CRE data vendors (CoStar, CoreLogic, etc.), manual research using Upwork consultants- Completed initial data architecture on AWS: S3, RDS (PostgreSQL), Redshift- Analyzed economic & financial data to select Metropolitan Statistical Areas for initial lending effortsWeb Applications- Managed development of several Consumer SaaS web tools for CRE lending (Quote Tool, Funding Tracker)- Gathered business requirements from originations / underwriting / marketing; managed web dev consultants (Five Talent Software)- Researched and selected tech stack: .NET Web API, React- Oversaw testing of web apps through usage of golden Excel spreadsheets, usertesting.com, optimizely.com A/B testing, etc.Management / IT- Helped hire Chief Data Scientist, Senior Database Developer- Interviewed and hired web app Software Architect; wrote job descriptions, created technical interview process- Wrote Engineering Department Culture document, implemented software dev tools (Atlassian) and best practices, created project plans / schedules- Established IT infrastructure (ISPs, firewall, internal network, standard PCs), hired IT firm -
Personal ProjectsCryptocurrencies Dec 2014 - Sep 2015Finished up some personal projects:- Researched cryptocurrencies: attended many Bitcoin conferences & meetups, read papers, watched Bitcoin videos, traded cryptocurrencies, studied microfinance- Learned how to program on the Bitcoin & Ethereum blockchains at Block Chain University https://www.linkedin.com/school/blockchain-university using early versions of Solidity and Remix IDE * Class project: https://github.com/huang-pan/village-bank
-
Lead Quantitative DeveloperT2Am Llc Jun 2011 - Nov 2014Los Angeles, UsT2AM is a hedge fund that is a leading authority in the field of algorithmic trading. Rishi Narang, the founder of T2AM, is the author of the book: https://www.amazon.com/Inside-Black-Box-Quantitative-Trading/dp/B08BLTR68FT2AM manages a portfolio of quantitative trading strategies (market neutral, short term trend, etc.) for Private Wealth Management firms. While at T2AM (a small firm), I created software tools that significantly improved the analytical capabilities of the firm. This directly resulted in higher returns for the firm.- Developed a suite of Portfolio Management & Risk Analytic web tools that improved the quality of the manager selection process. These tools included: * A Daily Return Analyzer tool that tracks the daily performance (VAMI, expected returns using Monte Carlo simulations, etc.) of each quant manager in the fund. Statistical analysis and data visualization was implemented using R and MS SQL Server in the backend. RackForms was used for the front end. The tool output reports in html / pdf format. * A Risk Factor Tool calculating the correlations of manager returns to various market risk factors (Euro, oil, gold, etc.) * A manager gross / net fee comparer in Excel / R that output an animated return histogram * R libraries: RODBC, dplyr, PerformanceAnalytics, parallel (CPU multi-core), ggplot2, googleVis, R Markdown- Back tested algorithmic trading strategies (e.g. turtle trading strategy) on various securities using market data (time series) from TradeStation and cross validation in MultiCharts- Applied the Pan Filter to create a lag less trend line trading strategy (superior to moving averages) in Matlab- Created a trade execution system in Excel VBA that interfaced with the Bloomberg EMSX API- Implemented software project management tools for the firm using Atlassian: Confluence, Jira, Bitbucket- Managed & supported IT infrastructure (Microsoft Sharepoint, Office365, etc.)- Developed technical interviews -
Professional Day Trader & Full Time StudentUc Berkeley Extension Online (Cupertino, Ca) Nov 2008 - Dec 2010Berkeley, Ca, UsFull Time Day Trader (Nov. 2008 to June 2010)- Investigated efficacy of a technical momentum based system trading the S&P 500 futures (ES) at eminiaddict.com- Gained knowledge of the markets & technical analysis; developed proper trading rules and risk / money management- Traded own accounts: most successful trade to date resulted in an 85% return with a 1% riskPrepared for UCLA Anderson Master of Financial Engineering program (July 2010 to Dec. 2010)- Passed CFA level 1 with highest marks in all categories- GRE 790 quant (92%) 700 verbal (97%)- Took prep. classes in statistics, economics, C++ -
Founder & CeoPan Filter Technology Sep 2006 - Oct 2008The Pan Filter is a revolutionary ideal filter developed by my father Dr. Cheh Pan. It solves the Gibb’s ringing problem that has plagued digital filters for over 100 years. The Pan Filter is a highly mathematical Digital Signal Processing algorithm based on the Fast Fourier Transform, and has been published and patented. https://www.researchgate.net/publication/3317921_Gibbs_phenomenon_removal_and_digital_filtering_directly_through_the_fast_Fourier_transform- Extended & improved upon the core technology / Intellectual Property through strong PhD level applied research- Developed real-time & data analysis software products for audio & image signal processing in Matlab; implemented the filter in C on a TI DSP- Looked at all areas of starting a company: created business, marketing, sales, finance, Intellectual Property, and legal plans- Presented the technology to different companies (including Matlab and TI) for licensing
-
Member Of Technical Staff Asic DesignRfmd Jun 2002 - Apr 2005Greensboro, Nc, UsWireless LAN department: formerly Resonext Communications, a wireless startup acquired by RFMD in Dec. 2002 for $133 million.- Lead an engineering team in the development & design of the PCI Express (PCIe) module of the Nepton chip * PCIe was the key interface that differentiated RFMD’s Nepton chip from its competitor’s products * Nepton was the first working PCIe wireless chip in the world, and was scheduled to generate $100's of millions- Managed a cross functional effort to establish PCIe compliance w/industry standards for Nepton chip- Successfully developed the Cipher coding encryption / decryption module of the Neptune 2 chip- Developed the USB 2.0 module of the Triton project; successfully taped out the Neptune chip- Evaluated new tools and improved the chip design methodology of the WLAN ASIC group -
Senior Digital Chip DesignerIospan Wireless Sep 2000 - Jun 2002Iospan was a wireless start-up founded by Professor Paulraj of Stanford University. I was instrumental in helping Iospan become the first company to successfully demonstrate 4G wireless in the world: its technology became the basis for 4G WiMax / LTE.- Designed & coded a wireless channel Interpolator (a poly-phase filter), a Symbol Demapper, and a Frequency Conversion Unit for one of Iospan’s MIMO OFDM broadband wireless chips (PHY: physical layer of OSI model)- Created / owned the PHY hardware spec; worked with marketing to convert the spec to a product datasheet- Completed the PHY ASIC top-level Verilog code & testbench; was the project verification lead for the PHY FPGA & ASIC- Helped prototype Iospan’s wireless technology on Xilinx FPGAs in the lab; helped test & measure the performance of the first PHY ASIC samples in the field- Required solid understanding of wireless fundamentals as well as the ability to work with circuit boards, the hardware / software interface (chip firmware), and lab equipment (oscilloscopes, spectrum analyzers, etc.) -
Digital Chip DesignerSun Microsystems Dec 1997 - Sep 2000Palo Alto, Ca, UsThe Graphics & Imaging department was responsible for creating high performance 3D graphics chips (GPUs) for Sun workstations. The FFB3 graphics chip was the largest chip at Sun at the time.- Lead a small team to design the Vertex Processor and 3D lighting unit of FFB3- Developed several patents during my tenure: U.S. Patents 5181-28100, 5181-89600, 5181-89800- Learned much about ASIC design from spec writing to rtl coding to synthesis & verification to place & route to chip tape out; Sun ASIC design methodology was the industry standard- Other duties included test writing for code coverage & formal verification, system & gate level debug, static timing analysis, and silicon test vector generation -
Junior Digital Chip DesignerCirrus Logic Jan 1996 - Dec 1997Austin, Tx, UsThe Entertainment Graphics department was responsible for creating 3D graphics chips (GPUs) for IBM compatible PCs.- Learned how to write chip design code in Verilog and how to run Synopsys / Cadence / Mentor Graphics ASIC design tools- Worked on 3D graphics chips including Microsoft's Talisman and Cirrus’s Magnum & Laguna chips; studied MPEG- Work included 3D geometry/triangle setup engine design, synthesis and initial place & route of a texture compression block, creation of a random test generator for a Rambus DRAM bus interface, and implementation of digital arithmetic structures (IEEE floating point adders, multipliers, reciprocators) -
Summer InternAmd Jun 1995 - Aug 1995Santa Clara, California, UsThe CAD Technology & Systems Division was responsible for creating tools that improved AMD’s chip design methodology.- Helped develop a Computer Aided Design (CAD) tool called Timing Budget Specification- Improved programming skills by coding a database syntax parser in C -
Research AssistantUc Berkeley Jan 1995 - May 1995Berkeley, Ca, Us- Worked with Chris Keller under Professor Roger T. Howe in the EECS department to design and fabricate polysilicon (MEMS) microstructures -
Summer InternSlac National Accelerator Laboratory Jun 1994 - Aug 1994Menlo Park, California, Us- Implemented a new Thermoluminescent Dosimetry (TLD) system on a Foxpro Database Management System for monitoring SLAC employee radiation levels- Trained technicians to use the technologically advanced system
Huang P. Skills
Huang P. Education Details
-
Stanford UniversityElectrical Engineering -
Blockchain UniversityLearned How To Program On The Bitcoin & Ethereum Blockchains Jan-Feb 2015 -
Ucla Anderson School Of ManagementQuantitative Finance -
Beijing Language And Culture UniversityIntensive Language Program -
University Of California, BerkeleyElectrical Engineering And Computer Science -
Saratoga High SchoolGd
Frequently Asked Questions about Huang P.
What company does Huang P. work for?
Huang P. works for Pond
What is Huang P.'s role at the current company?
Huang P.'s current role is Chief Data Officer.
What is Huang P.'s email address?
Huang P.'s email address is hu****@****hoo.com
What is Huang P.'s direct phone number?
Huang P.'s direct phone number is +140885*****
What schools did Huang P. attend?
Huang P. attended Stanford University, Blockchain University, Ucla Anderson School Of Management, Beijing Language And Culture University, University Of California, Berkeley, Saratoga High School.
What are some of Huang P.'s interests?
Huang P. has interest in New Ventures, Job Inquiries, Getting Back In Touch, Consulting Offers, Reference Requests, Career Opportunities, Expertise Requests, Business Deals.
What skills is Huang P. known for?
Huang P. has skills like Memory, Ibm, Snl, D3.js, Technical Analysis, Mimo, Perl, Computing, Jira, Mentoring, Nosql, Tracker.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial