Joseph (Jody) Winston Email & Phone Number

Houston, Texas, Us

Designing the next generation Managed Pressure Drilling (MPD) solutions starting with real-time data and ending with analytics.

Austin, Texas, Us

Designing the next generation of cloud native on demand insurance. Responsible for Engineering Best Practices. Led the organization by writing “six-pager” on Data Mesh.

Conroe, Tx, Us

Responsible for the design as well the implementation of the embedded distributed, soft real time application running on edge devices using ClojureScript.

Houston, Texas, Us

Delivering tomorrow's technology today.

Houston, Texas, Us

Transforming Halliburton’s services through the introduction of digital. Mentor for interns working on augmented reality. Tactical efforts include web platform architecture, university recruiting, and intern program. Strategic work includes mergers and acquisitions, data lake, real time, deep learning, as well as continuing education platform. Executive team member for Landmark’s 2019 Sales Kickoff with responsibility for Hackathon as well as Digital Bar. Executive team member for Landmark’s 2019 innovation forum with focus on OpenEarth’s Hackathon. Speaker at the University of Texas at Arlington's Fourth Annual Business Analytics Symposium on the topic of innovation.

Houston, Texas, Us

Building complex applications that cooperate on preassigned tasks. This includes “Blockchain” powered marketplaces for oilfield services such as wastewater disposal and proppant distribution as well as digital marketplaces for products such as deep learning models, data, and programs. Core team member for OpenEarth Developers’ Summit 2018.

Houston, Texas, Us

Delivering an Industrial Internet of Things (IIoT) architecture that enables Deep Learning at the edge of the oilfield. Tying everything together with Blockchain technology.

Houston, Texas, Us

Giving Data Scientists the ability to construct Geoscience specific workflows using deep learning opens opportunities to make new discoveries. I am providing the vision and the leadership to bring this transformation into the product suite.

Houston, Texas, Us

Big Data and Analytics have been part of the E&P industry since the introduction of machine aided signal processing. What has changed recently is that the rest of the world has rediscovered what we in E&P have always known. Converting data into information brings value. The innovation that the “cloud” architectures is simply running reliably at extreme scale on commodity hardware. My work has been to integrate these insights championed by Map/Reduce and Bigtable implementations on lamba architectures into the E&P space.

Houston, Texas, Us

I provide the vision for Landmark's software development as well as advocating for other software developers that use our products. While there is no such thing as a typical week when you are delivering a product as large as DecisionSpace, I spend time just about every day with the development leads and architects to keep watch over the progress. I meet with customers to understand what is working well and what parts of our products need improvement. I give our external customers the overview of the platform as well as the benefits that an integrated software architecture brings to the bottom line.

Houston, Texas, Us

I illustrate the potential that DecisionSpace brings to the E&P industry by showing others what is possible with this software platform. Adding NVIDIA’s IndeX, which brings scalable ray casting based rendering to the cluster, is but one example. Another project gave DecisionSpace the ability to use multi-touch devices like what Windows 7 provides. Pointing out the value of the ecosystem sometimes requires a dramatic change to adjust impressions. I took the entirety of DecisionSpace with the help of another and brought it to a new platform that was on the show floor at SEG. I understand the value of immediate feedback. I sped up the construction of GeoShells and gave the users the ability to edit these objects in real-time.

Sugar Land, Texas, Us

Moving a major application such as Landmark’s GeoProbe to a new platform is not an easy task. You need to bring all the software dependencies from one operating system to the other. This of course includes all the required libraries as well as development tools. I did the majority of this work for Landmark and built libraries for GeoProbe that included Open Inventor, CGAL, PNG, and TIFF. I wrote almost all of the CMake based build system for both platforms too. A new complier can point of questionable uses of code that other compilers might overlook. This is especially true when moving from Unix tools to those found on Windows. I worked through the code base and removed the majority of warnings found by Visual Studio.Removing technical debt is important in a long-lived project because there is a real cost associated with maintaining less than optimal code. I helped to remove problem areas by creating a set of generic geometry classes that replaced hand generated code. I also implemented an std::iostream logging system that replaced a home grown solution.Developing code is more than simply writing programs that work. True software development requires concern for the art of building software. I introduced the practice of online code reviews as a way for the entire team to see the proposed software changes and then to comment on the new code.

Houston, Texas, Us

Giving the end user the tools that they require to do their work as easily as possible is imperative as data sizes increase dramatically in the E&P industry. Sometimes this simply means reducing application development time so that the user can try out the program sooner rather than later. Using modern C++ development tools such as Boost helps to decrease the amount of boilerplate code and in turn brings a product out faster. I brought this tool to Shell’s and Bluware’s development practices. Adding a domain specific interpreter also helps. Not only did I contribute to the existing C++ based Python extensions but more importantly, I gave Petrel the ability to use Python inside of plug-ins. Another way to place software in the hands of the user is to have the user as part of the team. I successfully brought in domain experts into the group so we could have immediate feedback on the program under development.

Finding gaps in cellular coverage provides input to the design of new cell phone towers. I designed and implemented a system that took existing signal strength data gathered from diagnostic equipment and produced a map that showed where the signal was good and where it was not.Open source software has been slow to take off in avionics systems found in homebuilt planes. I created a glass cockpit that consumed data from different sources and produced virtual instruments.

Massy , Ile-De-France, Fr

Building next generation software requires supporting infrastructure. In order to construct a new seismic digitizing application as well as other internal products, I wrote tools that utilized the existing CORBA event service and exposed existing C++ and Fortran code using Python. I then applied portions of this framework to test and to extend the seismic processing data flow. One important extension gave the system the ability to track the processing history. Design patterns provide developers with a common vocabulary that describes what is happening. I gave other developers this language as well as using it in both new and existing software products.Legacy code is valuable because the results that the software produces have been verified over the software’s lifespan. Rewriting the algorithms can be costly as one implements the new version and compares it to the previous one. I understand the economic realities when maintaining extremely large legacy software verses re-implementing everything from scratch. Because of my experience, I successfully lobbied management to move our existing code base to a new hardware platform. This required me porting the entire seismic processing system to a new operating system and well as a new compilers.

Houston, Texas, Us

Moving an industry from desk-side applications to those that can run anywhere on the network requires vision to see what is on the horizon as well as the technical ability to navigate the emerging software landscape. I brought both to my client. I demonstrated the value of having a distributed log processing system by showing what is possible when one can have an application that runs in a browser that in turn can use data from multiple different data sources: Halliburton’s internal API, Landmark’s public API, and Schlumberger's well log file. This groundbreaking work enabled the remote monitoring of fields.

New York, Ny, Us

Knowing the capacity of a factory is a given in most business. With this knowledge, you have historical data on production as well as a good idea what the future holds. This information is valuable since it provides a good measure of the company’s worth. Imagine if you only knew what the past was but not what was coming next. This is often the case in reservoir management because it is difficult to compare the past to the present. I lead a team of five programmers that constructed a complete suite of tools that presented geo-scientists the changes of a reservoir over time. A key part of this work was the architecture of a distributed computational pipeline that orchestrated the data flow as well as the movement of the required algorithms. Simply dumping data about the changes is not very useful. Instead of doing this, I collected all the details about the reservoir in an electronic project notebook that enabled easy interaction with the information.

Houston, Texas, Us

The ability to extend a browser brings with it the exciting possibility of having one platform for software development. Using technologies from Netscape 2 and 3 such as the CGI extension mechanism, I gave users the ability to interact with well logs from distributed databases by writing plug-ins. Off the shelf CORBA technology made the movement of data from the remote server to the browser possible. This same technology stack for the wells does more than allow deployment in a browser. It also enables existing and new applications to run as is against new well log databases.

A mechanical tool that analyses pipeline corrosion and deformation requires an operator to record report. Numbers tell the information about what is wrong but actually seeing the changes with a simulated pipe make it easier to understand the damages. Using Iris GL to visualize the results along with a modern Motif interface makes this task easier. I made both possible by updating an existing application. Moving to a computer-based visualization requires a new platform with software to control it. I wrote the Linux device driver to control the camera that recorded the deformation and corrosion along with the driver for the stepper motors, which positioned both the camera so that it could read the damages. Simply using the current version of the hardware makes the software difficult to change, so I designed and implemented the classes that isolated the application as required. I used CORBA to move the objects that contained the deformation and corrosion objects over the network.

Rendering large seismic volumes in context gives interpreters insights that are difficult to achieve using other technologies. I brought Open Inventor experience to the original EarthCube team that brought the first version of the product to market in six months.

Houston, Texas, Us

Loading existing file formats into new applications allows users the ability to recoup their investment in earlier software products. I wrote software that performed this task. I also implemented APIs that allowed programmers to interpolate data.

Us

A user of a visual environment like what either Apple, Microsoft, or Google provide, increases productivity because it makes common, everyday tasks easier. This metaphor while apparent today was not always the case. One example drawn from my client who needed the ability to run many different Unix programs on X11, illustrates this point nicely. Users needed to quickly find their favorite programs. Developers wanted to know how their applications failed. Management required reports on usage that also included details on how often an application refused to run correctly. I accomplished all of this by leveraging a SQL database for the mandatory reports to management, by launching all the applications using Unix’s fork-exec semantics that allowed me to record trace backs on failure, and by giving the users the tools to find and run their selected applications.Adding coordinates by entering the actual values is tedious and error prone. I solved this problem by writing an X11 Input Extension that allowed the use of a CalComp 9500 Digitizer in all X11 applications.Researchers often want to use different languages to solve problems. I helped accomplish this by exposing C++ classes, specifically arrays, to Scheme programmers.

Houston, Texas, Us

Moving one of the world’s leading companies to a new operating system and a new way of doing business is something that only a few people have the privilege of doing. I was fortunate to help Royal Dutch Shell move from mainframes to personal workstations. Doing this required learning new languages like C with classes (Cfront), new user interfaces (SunTools, SunView, NeWS, X11), new visualization techniques (PHIGS, Iris GL), and new ways of sharing data (RPC, CORBA). I took this knowledge and worked in three different areas. One component of my time at Shell was the designing and implementing “work and view in 3D” programs for Basin Analysis. I created a way to view faults and other E&P data in three dimensions. I led the research on using Non Uniform Rational B-Splines as an effective way to display horizons and faults. The highlight of this area had to be producing a video of a prospect in the deep water Gulf of Mexico for the Royal Dutch Shell Board of Directors. I then spent time building tools. This work included real time device drivers for wire line data acquisition as well as PostScript divers for Versatec plotters. Most importantly, I contributed a large amount of code to the internal framework known as Spirit. The final are of work was writing a seismic quality control application. This was the first C++ application at Shell. My responsibility was the entire user interface, providing the C++ adaptors for Motif widgets, as well as the build system based on IMake.

Custom platforms built with hardware and software can enable new tools that never were feasible in earlier times. This is certainly true when you bring state of the art 80286 CPU and add in a high performance bit slice graphics processor. This cutting edge system gave us the ability to perform real time measurements on the human heart. I brought to the table the software expertise necessary to design the hard real time A/D acquisition system as well as streaming the observations onto an optical disk for retrieval at a later date. I modified the custom real time kernel and windowing system when needed. I also wrote the FDA 510(k) documentation.

The challenges of a start up are many. You need to refine the product's vision while remaining true to the idea's key insight. You must juggle many different roles and at the same time continue development at a rapid pace. You make decisions without enough information and yet must stay focused on the end goal. I know these are real tradeoffs because I was a software manager at a startup. While there, I set the direction of the software development from a group that suffered from the “not invented here” syndrome into one that saw the value of using off-the-shelf software. We built point of sale systems that also included all of the input devices. What set us apart from the competition was the fact we were the first to dedicate a general-purpose computer to this task in a cost sensitive industry. Keeping the overall price down meant using an in-house designed embedded system that allowed expansion through daughter boards. I selected the CPU and the overall architecture so we could use an operating system rather than rolling everything ourselves. This decision freed us from programming in assembly and gave us high-level tools that allowed me to build the industry’s first distributed point of sale systems. I created the application on top of well-known design patterns of using device drivers to provide an abstract interface. I designed the failsafe platform that allowed the application to transparently recover from power failures, which freed us from needing uninterruptible power supplies. I built the redundant distributed key-value database that allowed price lookups even when one or more systems went down. I even helped build the first five systems by wire wrapping the boards.

Houston, Texas, Us

Working with patients is an honor that makes one appreciate the human spirit. Inside of a clinic, I documented and analyzed abnormal human gait with the goal of recording a patient’s state so that one might see changes due to therapy or other treatments. This effort required the real time acquisition of how human muscle fatigues when given work. Electrical stimulation is one way to improve a person’s less than optimal gait. I designed stimulators and modified others to fit the patient’s needs.

Bachelor Of Science, Biomedical Engineering

Master Of Divinity, Postmodern Theology