• Grew a team from 1 to 6 engineers focused on building a data platform that enables stakeholders across Embark to gain access and insights into self-driving data to empower the development of the company's self-driving technology

• Worked with product team representatives and stakeholders across the engineering organization while balancing tech debt and team member goals to develop a strategy and plan that aligned to overall company OKRs

• Led team on pushes for ease of use of the data platform: abstractions around our internal databases with APIs, abstractions of our Airflow and Spark processing platform to empower ML and Data Science workflows, and faster data access formats

• Led design and co-implemented a novel data offloading system to support the company’s roll out of a commercializable product which enabled the selective offloading of vital data from the truck live during operation (Patents Pending)

• Redesigned and developed the data importing system used to offload terabytes of data into the cloud from trucks each day using Python, MySQL, and Amazon S3

• Redesigned and developed a data recording tool for the truck capable of handling 300 MBps of data using C++ and ROS

• Designed and developed a distributed data playback mechanism that enables engineers to test hours of self-driving data within minutes using ROS, Python, Docker, Kubernetes (Amazon EKS), RabbitMQ, and Amazon S3. Patent pending

• Designed and developed a low-latency video streaming for trucks in low LTE conditions using ROS, C++, GStreamer, Redis, and Python

• Assisted in the design and development of low-latency streaming of arbitrary data from autonomous vehicles in low LTE conditions using ROS, Python, and Redis (Patent Pending)

• Productionalized a teleoperation system for an autonomous trucks used as the book-end for Embark’s demos during its SPAC process using ROS, C++, Python, and GStreamer

• Redesigned and built our data access utilities to improve the speed by 10x using C++ (https://github.com/embarktrucks/embag)

• Built out a chunked uploading system to empower offloading data live from trucks to Azure Blob service in a reliable manner while operating in extremely unreliable network conditions using Python and Redis

• Built out a data request system that enabled users to select data points to be offloaded from the truck using the aforementioned uploading system. Used React, Python, and MySQL.

• Developed a simulation tool to determine if lack of human intervention would have resulted in a collision using C++ and ROS. Findings were published as part of Embark’s public 2018 Disengagement Report (https://medium.com/embark-trucks/embark-releases-our-2018-disengagement-report-2f242992141b)

Developed embedded software (C) and machine code used within memory interfaces on FPGAs, as well as Python, Perl, and TCL scripts for testing

• Designed and developed a coupon like system for user registration and management

using Node.js, Postgres, AWS Cognito, Amazon EC2, Docker, and Auth0

• Designed and Developed a statically hosted website to interact with the backend system

using React + Redux, Amazon S3, and Auth0

• Created various ETLs for a data warehouse using Node.js, Postgres, and RabbitMQ