AI innovation lab and startup studio
Full-stack software across an AI startup studio.
Verdant was an AI innovation lab and startup studio working across health, environmental data, legal technology, and other fields. I worked across the portfolio shown on its website, building full-stack software, backend services, and data workflows.
I built Python and React applications, APIs, databases, and background workers. The projects below show how that work connected language processing and graph models to interfaces people could use.
We used Jira and Trello for project planning and tracking.
In E-9000, I worked on the web workflow connecting language parsing, asset selection, and scene behaviors. Text became objects and actions that a person could inspect before assembling an editable Unity project.
The application connected a Flask and Jinja interface with spaCy and StanfordNLP, PostgreSQL, and Unity/C# scene logic. I implemented vignette screens, asset handling, and Unity/C# scene scripts.
I worked across the criteria-curation interface and backend views. The workflow let users review trial criteria, inspect extracted terms and medical concepts, and decide which concepts to use before matching cohorts.
The application used Python/Flask, Jinja, Bootstrap and jQuery, PostgreSQL, Redis/RQ, and Whoosh. I built criteria-review screens, medical-concept highlighting, and backend views.
Alongside the application, our team explored clinical knowledge graphs: connecting UMLS medical concepts and using those relationships to interpret trial criteria in context. That work included a Grakn schema experiment and research into weighted hypergraphs and graph-based matching.
I built an Apache NiFi workflow for TrialMix that retrieved database records, transformed Avro into JSON, extracted text, and sent it to MetaMap for medical-concept processing.
That work connected data ingestion and transformation with the clinical concepts used in criteria curation. It sat alongside the Flask application and its PostgreSQL, Redis/RQ, and Whoosh stack.
In a 2019 email-analysis prototype, I built a pipeline that turned unstructured messages into a relationship graph: who wrote to whom, and which people and organizations they mentioned.
Python and NLTK extracted the entities. I separated accounts, messages, people, and organizations, then generated typed nodes and relationships for CSV and Cypher imports. Redis/RQ distributed ingestion across workers.
The graph represented two different things: the communication between accounts, and the entities discussed inside each message. Keeping those relationships explicit made the text useful beyond a list of search results.
Another prototype let people assemble a decision process as connected questions, actions, and outputs. The Angular interface used a node canvas; the Python backend gave that visual structure a stored, executable form.
I implemented database models for nodes, edges, and rulesets, parsed the editor’s JSON, and added graph traversal and execution. Nodes retained their type, title, and canvas position; edges retained their source and destination. Execution began at a start node, followed question branches, and moved through linked actions.
This brought together interface design and backend logic: a diagram that people could edit, save, and run. The stack included Angular/TypeScript, jsPlumb, Flask, SQLAlchemy, and PostgreSQL.
I built device-data APIs and real-time biometric visualizations, including embeddable line charts. My work included the move to Python/Quart, device routes, chart templates, and WebSocket delivery.
The dashboard brought heart rate, heart-rate variability, respiration, temperature, and skin-response plots together. Vega-Lite rendered the charts; an overview plot let a viewer select a time window and inspect the signal in more detail.
The studio’s January 2021 reporting prototype compared actual performance with renewable-energy goals, then opened a quarter into location-level costs and consumption. Its React/TypeScript interface used Blueprint and Recharts.
I contributed to each of these projects as part of the studio team.
A system for translating medical text into virtual environments for psychotherapy and recovery under expert supervision. Language processing connects the text to the assets and scenes used to construct a virtual world.
View on Verdant (opens in a new tab)
Clinical-trial matching that connects physician notes with eligibility criteria. Natural-language processing and graph-based learning help researchers explore patient cohorts and identify suitable trial sites.
View on Verdant (opens in a new tab)
A framework for reusing engineering and business rules across data migrations. Genetic algorithms assemble modular transformations from a library of existing logic.
View on Verdant (opens in a new tab)
Environmental-impact data for investment research. Our team contributed natural-language processing and AI expertise to help wealth managers investigate the impact of specific assets and companies.
View on Verdant (opens in a new tab)
A marketplace for buying and selling biomass. Members can post supply listings or requests, negotiate directly, and move from an agreed term sheet to online contracting.
View on Verdant (opens in a new tab)
AI-assisted sales analysis for a specialized wholesale market. Decathlon supplied the models; our team at Verdant worked on making them stable, scalable, and repeatable.
View on Verdant (opens in a new tab)
Computer vision and automation for tracking hospital activity, equipment, and inventory. The project aimed to give operational teams a clearer view of resource use and staffing needs.
View on Verdant (opens in a new tab)
An econometric engine combining time-series analysis, ARIMA, Pareto analysis, and Shapley methods with data-platform integration. Built around the work of domain experts and analysts.
View on Verdant (opens in a new tab)
A medical IoT platform combining biomedical sensor signals into one view of a resident. The project explored how AI could connect fragmented readings to support care teams.
View on Verdant (opens in a new tab)
A confidential project connecting blockchain-based identity and document management with property-price analysis. Its scope included combining asset information and other data sources to estimate real-estate values.
View on Verdant (opens in a new tab)Project graphics: Verdant A.I.