pascal sturmfels

● Implemented a Spark Pipeline to simulate concurrent, high-intensity SQL queries in order to stress test existing client-facing SQL databases ● Performed a cost-benefit analysis of publishing data to SQL Server vs. querying cold-stored Parquet files from the cloud, and impacted management to adopt the cloud-based approach ● Designed a modular system to automatically monitor Azure SQL Databases and scale them down during periods of inactivity, reducing our organization's expenditure on said resources by up to 30%

Taught Theory of Computation to Computer Science undegraduates; I covered algorithm paradigms (greedy, divide and conquer, dynamic programming, graph algorithms), complexity classes and reductions (P, NP, NP-completeness), and design and analysis of algorithms.

● Developed a novel CNN architecture to predict age and cognitive decline from 3D structural brain imaging, which reduced per-patient loss by over 15% and reduced network training time by over 30% ● Developed the first robust, CNN-based pipeline to segment brains from fetal fMRI image; the pipeline is orders of magnitude faster than hand-labeling fetal images

* ● Developed a peer-to-peer communication app that is resilient to censorship and network blocking * http://1am-networks.org). * ● Designed and implemented protocols to simulate a mesh-network using the iOS Multipeer Connectivity framework. * ● Implemented persistent storage of multiple managed-object subclasses with relationhips using the iOS Core Data Framework. * ● Designed methods to keep mesh-networking alive in the background using the background execution framework.

* ● Designed a general online framework that improves the approximation bounds of existing algorithms for job scheduling in multiple settings. * ● Developed the first exponential-time algorithm to solve offline concurrent open shop optimally. * ● Implemented modern algorithms for coflow scheduling on randomized bipartite graphs in MATLAB to compare heuristic methods.

● Developed RNA-seq data visualization tools in R for the kallisto-sleuth software package (https://github.com/pachterlab/sleuth). ● Optimized the storage and retrieval of RNA-seq datasets in sleuth, compressing datasets enough for them to be hosted on a server. ● Built https://lair.berkeley.edu – a website to host interactive re-analysis of RNA-seq datasets, promoting transparency of published papers in genomic research.

● Designed a compact gas chromatograph to analyze compounds in biological systems. ● Optimized circuit layout of the gas chromatograph and implemented a PID controller for the gas chromatograph's temperature ramping.

● Designed circuits to monitor pressure and temperature of compressed air tanks, LightSail's form of energy storage. ● Analyzed sensor data in LabVIEW and MATLAB.

B.S.E in Computer Science, Minor in Mathematics Dean's List, University Honors, EECS Scholar, James B. Angell Scholar Activities and Societies: Director of Technology at Kappa Theta Pi – Professional Technology Fraternity