Tech lead for the Integrations team since I joined, which underpins two of three business lines, ingesting and parsing blockchain data from dozens of sources, and dozens more “smart” contracts

Most recently tech lead for customer-facing product, spanning big data storage, real-time and batch serving, and front end

Pioneered migration to linear project management, and use of Cursor AI-driven development, with MCP and other integrations

Assets service starting Q2 2024 - retired old APIs, fixed up retools, oversaw rewrite onto Iceberg, with 10x+ better throughput, and AI-driven asset acquisition, 10x’ing the size of our catalog, and introducing asset relations.

Enterprise Ingestion and Principle Markets Tooling Q4 2024, adding architecture docs, and full automation / self-service

Formulated our roadmap in consultation with cross functional partners to maximize opportunities for growth, enabling multi-millions in new recurring revenue opportunities. I also designed developed systems to enable auto-checking with the help of my team..

Led the team to make our systems 10-20 times faster and 100x cheaper to run, and we set the company standard for CI/CD, system transparency, and reliability, including auto-code-freezes, one-button rollback, comprehensive test suites and hooks with blue/green deployment

Improved the morale of the 9 person team through mentoring, vision, goal-setting, and pride in craft.

Organizational-wide contributions include driving internal data model liberalization, complete reorganization of our wiki pages, seeding a Tech Radar, standards on diagramming, tech design template, and “post-mortem” monthly discussions.

Worked as architect and chief of staff across all of Chegg Learning Services (seven offices, ~500 engineers, 30+ staff engineers directly) to develop joint standards and drive technical development and adoption.

* Moved from from <99.5% uptime across all systems in the company to 99.95%

* Ran Services Program (tech initiatives on the backend) with 20+ working groups at a time, developing: microservices (Java, Node, Lambda), database layer, inter-service communication with circuit-breaker, graphQL / federation, and API Gateway with AuthN/AuthZ; workflows with Kafka, Zeebe, and Step Functions, …

• Ran weekly international architecture leads meetings

* Ran >4/5-rated weekly engineering-wide learning sessions and retrospectives

* Ran annual internal summits with 45 sessions and external summit with >52k impressions, >2k likes, shares, comments, etc., 450 registrations, and 200 attendees, all averaging > 4.5/5 rating.

* Provided dozens of architecture reviews and support for all learning services teams and helped evaluate many new technologies (e.g. gRPC, Kotlin, PostGres) and vendors

• Standardized with working groups, RFCs, POCs, & adoption campaigns:

o Reference architecture for the full stack

o Project documentation & diagramming standards

o Data persistence solutions with Terraform: Dynamo + Aurora

o Data platform, from message bus through analytics & services

o API request and response standard parameters, headers, & fields

o Tech Rigor - multi-point audit process, and release checks

o Logging and application performance monitoring dashboards

o QE Standards

o US engineering hiring process, including interview training, onboarding & testing

o Training & employee mobility

* Worked with a vendor to develop an algorithm to automatically scale our container cluster in superior ways to what AWS provided.

* Product and tech leader for Data Catalog, maintaining all company data in AWS S3, indexed with MySQL, accessed through Java microservice, working with other Java microservices, powering Ruby UI.

* Leader of Individual Contributor Leads Roundtable.

* Developed architecture program and best-practices internal wiki-portal.

* Product leader for Data Catalog. Developed requirements and architecture. Led engineering team of 8 in Ukraine to build it. Presented and trained users. Exceeded target of 50 internal monthly users in first quarter. Core features: tagging and data extensions, duplicate prevention, auto-versioning, user query parsing, faceted search.

* Other systems architected and built: Feed Storage using Cassandra and JsonSchema, Receiving Java API + FTP/S3/MySQL storage, Code Critic webhooks patent-winning design, Access Rights authorization management UI

Responsible for converting print-PDFs from nearly every major textbook publisher into html5, keeping formatting intact and injecting semantics. Open-source solution saved company $5M annual contract, used to this day.

* Print margins (crop marks, annotations, and notes for the printer). I wrote a Hough transform in C to crop them out with 100% success rate.

* Extending pdftohtml, FontForge, node.js, and other open-source. MVP within three months, and 99.9% accuracy within a year. We traded the algorithms.

* Regularly reconvert, index, and QA 1-million title library: 600-node Condor cluster on volatile AWS spotinst cluster. Checkpointing and auto-restart. 200k lines of code and > 1M lines under management; almost maintenance-free.

* C/C++ system with Python test framework to achieve 80% reduction in the size and number of fonts; thirty point improvement in net-promoter score.

* Patented system to parse out figures and tables and enable pop-up content