Stephen D. Williams

Software / DevSecOps Engineer / Architect - Data - Robotics - Leadership

Full-stack Web, mobile, embedded, robotics, scalable cloud apps/APIs, security/crypto, startups

This: sdw.st/resume PDF: https://sdw.st/sdw2024.pdf

SF Bay Area / hybrid / remote 866-SDW-UNIX 650-450-8649 Fax:703-995-0407 github.com/sdwlig

sdw@lig.net stephendwilliams@gmail.com twitter.com/Scienteer LinkedIn.com/in/stephendwilliams

Highlights

Overall Experience

With a deep technical, creative, interpersonal, and business background, I comfortably fit a wide range of roles.
I am a generalist problem solver who has delved deeply into many technical & industry areas on many interesting projects:

Modern full stack web (React, WebCompnt, Mobile (Android), Database: transactional / data lake / time series / embedded,

Data architecture, codecs, semantics, standards, DevSecOps + highly secure application architecture & development,

Robotics / embedded (mechanical design & invention, CAD, 3D printing, embedded control, streaming video),

VR (WebXR Oculus app), Video streaming + codecs (custom WebRTC set top box, nVidia Jetson, video conferencing),

Deep awareness / study / prototyping AI/ML/Vision, visualization, UX design, legal (Patent Agent soon), Japanese,

Led teams up to 20 in large enterprise, government projects, large & small startups, founded several startups, non-profit.

Experience with electronics, fabrication, sourcing, branding & marketing, copyediting, private pilot flying.

Other experience includes: deep mobile dev (custom Android), cameras & imaging (MIPI, Movidius), ebook reader apps, scalable server apps for millions of users, secure app dev, insurance quoting, banking, encryption, PKI CAs, various databases, protocols & standards work (IETF, W3C), and embedded factory machine control. I’ve run my own servers with DNS, email, web, usenet, and apps on the Internet since 1992. Mobile Android (Java/C++) and desktop Qt (C++) programming, Java/C++/JNI, XML, web services, rendering. Vision recognition. Cray XT3 installation, repair, mgmt.

Created a new approach for robotic actuators (3 patents self-filed patent, 1 issued already); formed a large ANA Avatar XPRIZE contest team creating an avatar system: remote telerobotics for home healthcare & elder care, travel.
Working with Linux, ROS, ROS2, WebRTC, WebSockets, Oculus Quest, mechanical design.

Experience

RTX AST-Raytheon Technologies / Collins Aerospace Applied Signals Technology San Jose, CA, 4/2023 To Present

Working with network communications, Neo4j graph database, TimeScale time series database, Java, C++, Javascript, ElasticSearch database, KVM, Kafka, NNG, AWS administration & development. Ramping up on AI/ML LLM, generative, vision technologies.

RedPillVR Los Angeles, CA 4/2023 To Present

Played a key role in solving the remote Internet streaming stereo VR from a complex Unreal Engine game to a VR headset like the Meta Quest 2 & 3, via both a pure WebXR web application with embedded WebAssembly compilation of complex multi-threaded C++ video codec and a native Android C++ application. Solved connection orchestration, streaming pose + stereo frame management for low latency, achieved Asynchronous Time Warp (ATW), first through custom solution then via OpenXR layer support. This involved complex debugging of several layers of frame processing on both sending & receiving side, modifying Unreal Engine internals, on both the C++ / Unreal side, and the C++ / WebXR Meta Quest 3 side.

WebXR, Android, WebAssembly, WebRTC, HTTP3/QUIC/WebTransport + creating generalized messaging layer, Unreal Engine application development, frame streaming integration, OpenXR / LibOVR custom remote VR streaming, HEVC + LCEVC video encoding / decoding, low latency streaming management protocols, Go orchestration server. GPU shader program development & debugging on OpenGL / Android / Quest VR environment, UE DirectX12 Windows environment.

Ramping up on AI/ML LLM, generative, vision, robotic control technologies.

Exploratory projects Sunnyvale, CA / Vallejo, CA 4/2020 To Present

WebXR, Android, React, React Native, WebRTC, Google Cloud Platform (GCP), TileDB.

Linux, Oculus Quest app, ROS, ROS2, CANbus, WebRTC, websockets, Ortelio cloud robotics, C++, Python, JS.

Mostly ready to be certified as a patent agent. AR/VR/RR. Trained on Udacity’s robotics curriculum.

Chatbots, NLP, translation, GAN, video scene understanding.

Toyota - Woven Planet, Renovo Motors - Senior SW Engineer - UI / full stack 2.5 yrs Palo Alto 9/2020-2/2023

Supported AV / ADAS / ECU development support work through time series & data lake database query & visualization app, including video + radar scrub.

Led UI team rewriting & extending management & visualization UI for ADAS (advanced drivers assistance systems - self-driving / autonomous driving technology) research support. React, webpack, lerna, GraphQL, WebSockets, GoLang, AWS, Docker, Kubernetes, Grafana, Postgres / RDS, Timescale, Influx, NATS, MQTT. Providing other technical architecture & detailed support as needed: security architecture, scalability, Linux, embedded.

Investigated HTTP/3, WebTransport, communications visualization, bazel, C++, rust, flutter, R-Car.

Yebo Technologies, formerly: Change My Path - CTO/founder San Francisco, CA 5.5 yrs 10/2014 To 04/2020

https://helloyebo.com/ Demo video: https://sdw.st/y.mp4 https://cped.co/d

Creating a scalable self-reinforcing social education platform, entertainment with learning to solve the job skills training gap, meta-learning deficiencies, and later to revolutionize life-long learning in general. Sourced near-shore team members in Mexico & Honduras. Negotiated & closed sales, pitched to investors, board meetings.

For several years, I have been building an ambitious single page web app, a complex integration of 3D game engine (WebGL, ThreeJS), 3D editor, video, rich text editing, slides, and scripting in a highly responsive web app using web components. Several tools published as open source: Merkle Hash web cache builder, completely new panel responsiveness system, async module loading system, Docker, websockets, graph database based app stack.

Deep Javascript, web dev, CSS, full-stack, webGL, ThreeJS, WebComponents, Polymer, Lit-Element/Lit-HTML, haproxy, nginx, docker, dgraph, NodeJS, AWS, Auth0, Cloudinary.

Instructional design R&D, novel theory of knowledge - concept morphology, invented “panel responsiveness”, Merkle hash web cache system, no-build web module load system, anti-browser memory leak techniques.

Pelican Imaging - Director Software Engineering Mountain View, CA 2.5 yrs 4/2012 To 10/2014

Building embedded software and SDK for revolutionary depth array camera technology for mobile devices.

Screened and interviewed hundreds of candidates. Developed a particularly revealing Socratic interview problem.
Solved numerous management, product, and technical problems while managing and working with technical teams, direct and offshore. Acted in management, product, team lead, and deep hands-on development roles.
Suggested and drove key product, product marketing, optimization, imaging, computer vision, technology, and methodology ideas, features, and implementation details. Overseeing IT, saved approximately $100,000.
C++11, CMake, apps & dev for Android / web / Qt - MacOS / Qt - Windows. Movidius Myriad 1, JavaGlue, Java

GREE / OpenFeint - Principal Android Engineer Burlingame/San Francisco, CA / Tokyo 10/2011 To 3/2012

Built new generation social gaming and casual gaming platform API for Android.

Building concise API for game developers use of GREE / OpenFeint social and casual game support features.
Support Android, mobile (iOS), and scalable server architecture and development environment and process.
Implementing OAuth, JSON, image acquisition, and other capabilities on Android.

Sony Electronics - Consultant San Jose, CA contract max 1.5yrs 5/2010 To 9/2011

Mobile & desktop eReader - Android, Qt/C++, OAUTH, XML, Adobe RMSDK, DRM, Java, C++, Eclipse, CMake

Heavy problem solving, developed new eReader app for Android in Java & C++ and desktop in Qt and C++.
Reading view: 2 page mode, RtoL, zoom/pan, page-flip anim., prerender, font swap, bookmarks, and highlights.
Implemented all of OAUTH, multi-threaded apps, DRM, Android, Qt applications.
Created a DOM-like XML parser in Java from scratch with a Java-optimal API, built-in SAX parser.
Forked the XBig Java->C++ integration code generator to become the JavaGlue project.
CMake system: MacOS/Linux/Windows/Android, C++/Java/Java->C++/C++->Java, multi-project..

High Performance Technologies, Inc. - Senior Technical Director Arlington, Reston, VA ~8yrs. 10/2001 To 1/2009

Development, security, and consulting for 5 main agencies / projects (US Treasury, HHS, DOJ, DoD ATL, unnamed, DoD HPC), several additional exploratory projects and agencies.

Active and successful in architecture, development, consulting, management, business development, proposal writing, market and technology research & evaluation, technical interviewing, and program assessment. Projects:
Army supercomputers, temporal graph analysis, W3C Efficient XML Interchange, DoD Acquisition review team
DOJ CIO PKI/Security, HHS PKI CA, Treasury FMS Secure Payment System
Qt/C++, Java, Javascript, Linux, HPC / supercomputers (Cray)

Previous experience:

Concinnous Consulting, Inc. - Owner / Senior Consultant Ashburn, VA 8 yrs. 2/1992 To 9/2001

Geico, Chief Architect Geico's first viable web auto insurance quoting system. MPEG4 File Format encoder.
Contracts:

America Online, Inc., Senior Consultant Tyson’s Corner, Reston, Dulles, VA 3 yrs. 4/1995 To 5/1998

Developed Buddy List. Part of Smithsonian permanent collection. Created video IM system - Instant Images, first for large video IM service. Digital notary, rate-based protocol video streams.

Bank of America, Consultant Firewall and internal and external web servers - first public web server for any bank.

Lexis-Nexis, Consultant 3 yrs. Feed search engine in C++, RDBMS. Stock quotes in C++, satellite data reception.

Jabber.com, CTO Enhanced features, roadmap for extending Jabber's open IM/Presence system.

Insta.com, Founder Fusion of presence, instant messaging, dynamic web pages, and customer service.

ReverseAuction.com, CTO Architected auction web application, resulting in successful release, scaleup.

GE Aircraft Engines 2.5 yrs. Distributed process diagnostic, recovery. Ported g++, Emacs, Kerberos to HPUX.

Computer Sciences Corp. 1 yr Created language, forms in C++ much like today's HTML Web forms, inline scripting.

GE Lighting 1.5 yr Embedded control bulb machine. New class of bulb: half of the re-lighting Statue of Liberty in 1986.

Education & Training

American University - Master of Science, Computer Science - 4.0 GPA (completed 2013)
Stanford online: AI: Thrun/Norvig, Machine Learning: Ng; audited: Robotics, Graph Reasoning, Discrete Math
CISSP 54017, NSA INFOSEC Assessments, Lifetime AAAI. FBI Infragard. IETF, W3C, Certified Scrum Master

Innovation Highlights

2022: Self-filed a new robotic joint approach combining low cost / noise / weight with strength, simplicity, 3D printable, few parts, compact support of multiple joints.

Patent published SPIVT robotic actuator: filed pro se. Patentable novel web panel responsive algorithm.
Patent issued for image to texture in 3D games, contributor. Patent filed on May 2010 eReader features.
Independently invented CAPTCHA in 1999. Created IM-based multi-person video chat system at AOL.
Wrote & scaled AOL's Buddylist, inventing independent claims in AOL’s Buddylist patent US7454709.
Founded early ISP, have run Internet servers continually since. Implemented satellite delivery of Usenet.

Examples

github.com/sdwlig - W3C XBC Measurements Methodology - W3C EXI Measurements Note - ESXML

I run multiple half marathons and other races every year, biking, mountain hiking, kayaking, sailing, and inline skating. I have run at least 20,000 miles. I am a private pilot, PADI scuba certified, rock climb, ski, and snowboard. I’m an avid semi-pro photographer: over 100,000 photos taken globally.

One of my publishing venues:

https://medium.com/missed-points https://medium.com/optimalogic