Erick Shepherd

← All skills

Skill

Python

Related work and learning, in one place.

Experience

  • Application Development Advisor (Full-stack software engineering)

    Cigna Supplemental Health Services (via Magnit Global) · Aug 2021 – May 2026

    Sole developer maintaining and extending an inherited field-quoting application and lead on the contract-generation app behind a national supplemental-health-insurance underwriting pipeline — Oracle APEX and JavaScript on the front, PL/SQL and Python on the back.

  • Systems and Embedded Software Engineer

    Self-employed (part-time) · Dec 2020 – Apr 2024

    Independent contract engineering spanning embedded/IoT devices, data-analysis pipelines, and web systems.

  • Research Intern

    NASA Goddard Space Flight Center · Jun 2019 – Dec 2019

    Streamlined ground-truth validation and eliminated a proprietary IDL dependency in NASA's MODIS Dark Target aerosol-retrieval pipeline; two tools filed as NASA New Technology Reports.

  • Research Intern

    NASA Goddard Space Flight Center · Jun 2018 – Aug 2018

    Studied poleward aerosol transport using space-based lidar from the International Space Station.

  • Research Assistant

    UMBC — Department of Physics, ACROS Group · Aug 2017 – Jun 2020

    Atmospheric remote-sensing research on UMBC’s Taki supercomputer. NSF CyberTraining research in fall 2019 and coursework in spring 2020 were part of this longer appointment, with stipend support.

  • Teaching Assistant, Computational Physics

    UMBC — Department of Physics · Jan 2019 – May 2019

    Spring 2019 appointment for Introduction to Computational Physics (PHYS 220).

  • Teaching Assistant, Computational Physics

    UMBC — Department of Physics · Jan 2018 – May 2018

    Spring 2018 appointment for Introduction to Computational Physics (PHYS 220).

  • Co-Founder and Chief Operations Officer (Self-employed)

    Industrial Commons, Inc. · Sep 2015 – Dec 2017

    Co-founded and ran operations for a nonprofit community makerspace.

  • Computational physics teaching materials

    My 2018 computational-physics teaching solutions implement trapezoidal and Simpson integration, bisection and Newton root finding, interpolation and linear regression in Python.

Projects

  • OracleDB AI Bridge — An Internal Developer Tool

    Project

    I built a local Python utility that colleagues used to query Oracle databases through Claude and Cursor. It supported database and dependency analysis across more than 15 applications.

  • Earth-Science Data Tooling — Packaged Scientific Python

    Project

    I developed scientific-data tools through NASA internships and UMBC research, including conversion to netCDF4 and parsing into pandas DataFrames. cosmic-crunch and noaa-gml-file-reader are published on PyPI; I also contributed to collaborative cloud-anomaly analysis.

  • ncarnate — Legacy Scientific Data to Modern netCDF4

    Project

    I rebuilt ncarnate to convert supported HDF4 and netCDF data into netCDF4, verify stored values against the source, and reconstruct coordinates for supported satellite grids and swaths. The Python CLI and library are published on PyPI and conda-forge.

  • Gestalt — A Multi-Provider AI Workspace

    Project

    I architect and direct Gestalt’s AI-assisted development: a workspace for conversations, prompt skills, and provider execution. Its Python, React, and Rust components separate prompt content from permissions and connect to tools through MCP.

  • excubitor — Safety Fences for Autonomous Coding Agents

    Project

    I built excubitor to enforce Git checks in unattended Claude Code workflows. Python hooks block specified actions outside the model, while a separate audit checks project claims against executable evidence.

  • did-it — Checking an AI Agent's Claims Against Its Transcript

    Project

    I built did-it to compare a coding agent’s procedural claims with its Claude Code transcript. It checks the sequence of tool calls and results, then returns a receipt identifying supporting, contradictory, or missing evidence for each claim.

  • Coding-Agent Evaluation Harness

    Project

    I developed an AI-assisted Python harness to compare coding-agent results against acceptance checks fixed before each run. Recorded provider runs and test results support repeatable comparisons on the selected tasks.

  • LLM Guidance Routing and Evaluation

    Project

    I developed a deterministic Python router for prompt guidance with AI-assisted implementation. It resolves dependencies and context budgets; controlled comparisons tested whether structured revision improved results over the baseline.

  • Jarvis — A Voice Assistant

    Project

    I architected and directed Jarvis with AI-assisted implementation, building a working voice assistant that connected wake-word detection, speech recognition, pluggable language models, MCP tools, and speech synthesis.

  • Walter — An Experimental Virtual Assistant

    Project

    I built Walter as an experimental virtual assistant before the widespread use of large language models. It combined listening, speech recognition, punctuation restoration, and synthesized cues, alongside experiments in language parsing and facial recognition.

  • voice-training — A Local TTS Fine-Tuning Pipeline

    Project

    I developed an AI-assisted Python pipeline to prepare audio and fine-tune XTTS-v2 locally for personal, non-commercial use. It produced a versioned voice bundle with a pinned synthesis configuration for use by a separate assistant.

  • pydlock — A File-Encryption Package on PyPI

    Project

    I rebuilt pydlock from a file-encryption script into a Python CLI and library published on PyPI. It combines scrypt password-derived keys, Fernet authenticated encryption, and a versioned file format.

  • Monte Carlo Photon Transport

    Project

    I built a Monte Carlo photon-transport model to explore how a cloud's optical thickness and absorption affect the light passing through it. The project began as my computational physics midterm at UMBC in spring 2017; the browser explorer here reconstructs that problem.

  • Service Operations Map

    Project

    I turned a septic-services business’s customer spreadsheet into a touchscreen map with pricing zones, service-due heatmaps, and upsell flags. The shop adopted it in place of its sticker wall map.

  • Language and editor tooling

    A custom language with a lexer, parser, scope checks and Python AST execution, plus TypeScript editor extensions, a language server and a debug adapter for diagnostics, source breakpoints and stack frames. I architected and directed the AI-assisted implementation.

  • Smart-contract development

    Solidity smart-contract work includes token systems, controller and timelock logic, and Python web3 automation.

  • Network and audio experiments

    Python experiments with scapy and nmap for network reconnaissance and packet inspection, plus audio-based information hiding and signal processing. I also used Kali Linux and analyzed public password-breach datasets in educational work on owned systems or with consent.

  • Rust scientific-format parser

    I architected and directed a pre-alpha Rust parser for a documented read-only HDF4 subset, with PyO3/Maturin Python bindings, bounded fuzz targets and differential comparisons. It does not claim complete HDF4 compatibility.

  • Service-management web application

    An unfinished pro-bono Django prototype uses relational models, database migrations, forms and URL-routed HTTP views, with SQLite during development.

  • Source-derived architecture diagrams

    I directed development of a Python tool that parses source ASTs, extracts dependency relationships and generates Mermaid diagrams, with a comparison check for stale diagrams.

  • Game-authoring and 3D asset pipelines

    I architected and directed a validated DSL-to-Godot authoring pipeline and Python/Blender processing for generated 3D assets. The prototypes use GDScript, procedural geometry and content-addressed artifact storage.

  • Python interfaces and web extraction

    Tkinter canvas animation and keyboard controls, plus BeautifulSoup pagination and structured extraction in Python utilities.

  • Scientific C/Python wrapper

    A Cython wrapper exposes an existing C scientific routine to Python. The contribution is the integration layer, not authorship of the underlying algorithm.

Education

  • Electronics and measurement at UMBC

    In Electronics for Scientists in Spring 2021, I completed CircuitLab simulations and worked in paired labs on passive and active filters, op-amp and timer circuits, and digital logic. My reports include circuit calculations, measured data, oscilloscope captures, error propagation, truth tables and flip-flop results. My homework fits exponential and sinusoidal models with SciPy and plots measurement uncertainties.

  • Computational physics at UMBC

    I completed computational physics in 2017, before returning as a teaching assistant in 2018 and 2019.

Writing

Courses & credentials