After two great years with FireEye — and working with some of the best engineers in the industry — I decided that it was time for me to step aside. As of this month, I’m back on my own again.
So what's next? Well, I'm pursing a handful of fun projects. First, I’m shifting my focus to IoT; more specifically, applying sensors, drones, and robotics to precision viticulture. I live in Albemarle county, Virginia — a beautiful area of the country with an emerging wine industry. Fortunately, I live next to a prominent vineyard in the Virginia Monticello wine region. Over the last couple of years I’ve had an opportunity to look over their shoulders and learn about the pleasures and challenges of operating a vineyard. From what I’ve gathered, I believe there’s a huge opportunity to apply sensor and data science technologies to viticulture. Therefore, I’m motivated to develop solutions that local growers can use to optimize their operations.
Secondly, I’ve always been and still am a champion of Suricata. I consider Suricata to be one of the most powerful threat detection engines available to the public. Moreover, I love working with the core team and interacting with the international community. During the next few months I will be immersing myself in the development of a few features. My first project will be implementing a packet acquisition module for DPDK. Why? Because over the years, the DPDK project has matured. Intel is investing additional resources and pushing to make this the de facto library for high-performance packet processing. Given that majority of the Suricata deployments are hosted on Intel-based commodity servers, it stands to reason that Suricata should have support for DPDK.
In addition (this ties back to the previous paragraph), I believe that the Suricata engine is suited for monitoring IoT networks as well. On my farm, I have two custom security gateways running Suricata on Raspberry Pi 2. This is a great example of the versatility of Suricata. It can run on powerful big iron and small constrained single-board computers (SBC) with no code changes. When it comes to IoT, I’m curious to see what kind of data is flowing out of devices deployed throughout the farm. However, this is not yet possible until there is support for IoT-specific protocols. I am aware of an effort underway to implement support for link-layer protocols likeZigbee or 6LowPAN. However, to complete the solution, there needs to be support for high level protocols like MQTT and CoAP.
Finally, there are plenty of other side projects that have my attention. For example, I have some ideas about remote sensing for viticulture using drones for collection and machine learning to analyze the imagery. Plus, I'm working with a group of veterans to design, implement, and open-source a growing system for microgreens using a recycled shipping container, hydroponics, and Raspberry pi-based control system for automation.
All in all, I believe the next wave of innovation -- especially for agriculture -- is in the direction of IoT (sensing) and machine learning (analyzing), and robotics (actuating). Very exciting stuff!