Staff Software Engineer on Mobile Enablement Team. Worked across the full stack, from mobile data capture in Rust to server-side data pipelines in Go, in addition to UI work in TypeScript. Paid down long standing technical debt through my own initiatives across many languages (Go, Rust, TypeScript, Java, Objective C) and many build systems (Bazel, Groovy/Android, CMake). Gave demos of our team's work during our regular show-and-tells, and also presented on my own topics of interest at our monthly lunch-and-learn.

I am a member of the Data Science team and I report directly to Manyam Mallela, the Chief AI Officer & Co-Founder.

As a Principal Software Engineer my insight and expertise are called upon to identify, communicate and execute new engineering initiatives. I take point on challenging engineering problems such as architectural design, feature implementation, and the scaling of our services (in terms of both throughput and stability) to meet our performance targets; whenever possible, I delegate and provide guidance to promote growth in the team. I also share my knowledge and architecture reviews/postmortems through internal talks and written documentation.

When I joined, I took ownership of the core product recommendation service (written in Rust) that feeds our customers' marketing campaigns, responsible for each product of every marketing email/sms/etc that is ultimately sent out. Work primarily consists of optimization and feature enhancement of the interpreter (our customers' recommendation schemes are expressed as code written in Plasma, an interpreted language of our own design). By capturing new metrics and performing additional profiling, I made large reductions to memory use, significantly increasing concurrent throughput -- ultimately allowing us to use smaller EC2 instances, thus reducing our compute expense. I rewrote the parser (originally written in Ruby) and overhauled the interpreter so we could provide the precise line/column where errors occurred, allowing us to more quickly respond to any issues. I would be happy to expound on profiling, optimization, availability, caching and other topics relating to my work.

I additionally contribute to other systems where my expertise in debugging, systems architecture and implementation could be best leveraged.

A few of the technologies used: Rust, Ruby, Elixir, Kafka, Pulsar, AWS.

Leading design and implementation of core security and HTTP/gRPC proxy components within our service mesh:

Designed, documented and currently managing a team responsible for the roll-out of Zero Trust Networking infrastructure. We are implementing SPIFFE (Secure Production Identity Framework for Everyone), leveraging SPIRE within our Kubernetes/OpenShift clusters for X.509 key issuance/rotation. Lyft's Envoy L7 proxy is used for TLS termination in a sidecar configuration, using its SDS API to push certificates to each service's respective sidecar. All service-to-service communication is secured (in terms of both authentication and authorization) via mTLS. Intra-mesh communication is subject to RBAC policies based on the service's SPIFFE-ID.

Designed and implemented an extended version of the Envoy proxy, empowering our developers to write Envoy extensions (e.g. filters) entirely in Go, whereas vanilla Envoy requires writing C++. Implemented with C++ and Go, using Google's Bazel as our build system. As a testament to our ingenuity, Google has never wielded this confluence of technologies, as evidenced by our need to work around showstopping bugs in their software, which, 5 years since reporting, have yet to be resolved: https://github.com/bazelbuild/rules_go/issues/1486

As a fully remote employee, I leverage teleconferencing, team chat and shared documents to communicate my designs/specifications and tasking.

As employee #5, I've worn many hats, including mentoring of interns, profound performance improvements to our Continuous Integration infrastructure, government contract work, and much more. I'd be more than happy to go into details if asked.

Design and implementation of our email infrastructure, written in Clojure. The system consists of a REST web service (for enqueueing email jobs and providing reports) and a high performance, concurrent mail injector (which works the jobs: rendering email templates, personalization, submission to our Mail Transfer Agent, etc).

Also providing support and guidance for our internal email tools, which are Ruby/Rails applications that are clients of the aforementioned infrastructure.

Design and development of the iFly 700 software - GPS moving map software for pilots (WinForms application on Compact Framework 3.5).

I implemented real time weather rendering via ADSB signal, pilot/elevation obstruction alerts, METAR parsing, etc. Check out the product description for more feature info; if you happen to be a pilot, and don't one an iFly, buy one. Now. It's really that awesome.