Poul Costinsky

Seattle, WA.

Skills:

• Strategic planning
• Product development
• Team management and building  Mentoring
• Full software development cycle
• Jump-starting big projects
• Machine Learning
• Artificial intelligence
• Cloud architecture
• Data processing platform design
• Business continuity processes
• Architecture
• Software Design
• Software Development
• Cloud Computing
• Big Data
• Knowledge Discovery
• Data Mining
• Online Security
• Internet Safety
• Scalability
• Parallelism
• Distributed Algorithms
• SPARK
• Scala
• Cassandra
• NoSQL
• .NET
• Web Services
• Databases
• Windows Azure
• Database Development
• Algorithm Design
• OOP
• System programming
• C#, C++, SCOPE, SQL, Javascript

Experience:

Facebook 2020-current

Role: Principal engineer, tech lead. Business Integrity, Actor Integrity, Privacy.

Truthset.io 2019-2020

Role: CTO, Cofounder in a startup.

• Strategic planning, product development, funding, competition research, market research, analysis of potential customers and data partners.
• Building tech stack, data science, and back end.
• Hiring and mentoring

TheTradeDesk.com 2017-2019

Through acquisition of Adbrain

Role: Principal Software Engineer, AI, Machine Learning

• Working across Data Engineering and Data Science Teams.
• Driving adoption of AI and Big Data through shipping high-priority products.
• AI Guild – building quick AI prototypes in collaboration with data scientists and engineers from various teams, both for education and as a proof of concept for planning of major releases.
• Strategic planning and investigations in response to changes in programmatic landscape.

Role: Principal Software Engineer, Scrum master. Reporting to SVP of Engineering.

• Managing transition of the Adbrain team into TheTradeDesk.com post-acquisition.
• Same roles as in Adbrain.com, plus:
• Managing integration with the main platform;
• Sharing Big Data and ML/AI experience and knowledge with the rest of the company;
• Onboarding and mentoring new team members.

Adbrain.com 2016-2017

Roles: Chief Software Architect

• International startup with 30+ employees.
• Identity, AdTech and MarTech software – Identity, Person, Household multi-layered graph of addressable users on Internet.
• Navigating through several shipping cycles, to the sale of the company.
• Product definition, planning, scheduling, shipping and post-shipping software life cycle.
• Architecture, design and implementation of Big Data (Petabyte scale), Machine Learning (no, really), Artificial Intelligence .
• Sticking to a very limited budget while achieving scale 12x.
• Privacy, GDPR, compliance.
• Scale, performance, maintainability, manageability, governance.
• Mentoring teams of Engineers and Data Scientists.
• Running Scrum.
• Developing algorithms.
• Writing, testing and shipping code.

Technologies:

• AWS – S3, EMR, Data Pipelines, Lambda.
• SPARK.
• Scala.
• Spark ML.
• Random Forest, Logistic Regression, Linear Regression, Graphs.
• Java, Beanstalk, Tomcat.
• JIRA, Bitbucket, Git, IntelliJ.

Microsoft 1998-2016

Roles:

• Principal Software Architect.
• Principal Software Design Engineer.
• Architecture, design, implementation, stabilization, shipping, monitoring and auto-recovery of entire products.
• Writing code.
• Coordination between teams and divisions.
• Design reviews.
• Mentoring.
• Pilot projects.
• Writing specs.
• Overseeing spec writing.
• Project and team troubleshooting.
• Code reviews.

Previous Projects:

• Internet Reputation store with Business Protection Layer.
  • Porting SQL-based reputation store to NoSQL. o Adding new features, scale up, and tuning performance.
  • Testing and validation of the functional equivalency on the live system protecting 100s of millions of users.
• SploitStopper – client-server-backend solution to mitigate browser exploits. o Architecture of the end-to-end solution: NoSQL back end, front end servers, client side caching, synchronous reputation lookup.
  • Design of the client-server communication protocol.
  • Implementation of the back end.
    • Serve the business need for the immediate project.
    • Extensibility to other entities.
    • Goal to reduce setup time of a new system from months to days.
• Windows Online Store Safety.
  • Jump-starting the project. o Identifying threats and mitigation strategies. o Working with researchers to create the whole reputation system, driven by machine learning. o Architecture, design, and implementation of NoSQL backend of reputation-based safety intelligence system.
  • Helping partners to onboard NoSQL. o Shipping reputation data to the partners in the Windows Store.
• Family Safety.
  • Porting SQL based reporting service to Azure Storage.
  • Porting front ends to Azure hosting.
• Applications reputation – mitigation of malware distribution.
  • Creating “White list” reputation algorithm for executable downloads.
  • Architecture, design, and implementation of NoSQL backend. o Backend management of machine learning algorithm outputs.
• Malware harness
  • Architecture, design, and development of the system that stealthily downloads executable from internet and scans them for viruses.
• Smart Screen – anti-malware, anti-phishing client-server-backend solution for IE.
• Internet Browsing Safety.
• Data Intelligence.
• Internet Data Gathering.
• Virtual environments.
• App/cloud integration, early 2000s.
• Accounting software for small businesses.
• Business Contact Manager.
• Business Central online portal, Web 1.0.
• SOAP protocol development.
• Pub/Sub, Messaging architecture in .Net pilot.
• Messaging bus in WMI.

Hashavshevet (Israel) 1990-1998

Software design engineer

• Data Mining.
• Algorithms.
• Imaging.
• Structured data exchange.
• Document archiving.
• Accounting.
• Knowledge Discovery.
• Online Security.

Kineskop (Ukraine) 1987-1990

• Software design engineer.
• Hardware design engineer.

Education:

Moskovskii Institut Stali i Splavov

1982 – 1987 M.Sc. Computer Science/Applied Math/Metallurgical Engineering Automation.