Welcome! Let's talk about Overseed
I founded Overseed out of the frustration I experienced in creating, managing, and sharing simulation data. We use simulation data to demo applications, test systems, or train algorithms when the data is not available.
I've worked in the data space for over a decade. I served as a technical resource in various industries, including financial services, insurance, and retail. Throughout many projects, I found data design to be a primary factor in an application's success. Ultimately, data design is the creation of a model and its use throughout the development life cycle. In most projects, the process was as follows:
- Design a data model
- Generate data for the model
- Build a prototype
- Deploy the application
In all of these projects, I noticed the team spent a significant amount of time building data models and maintaining data generators for these models. The tools that supported this process were fragmented or non-existent. This resulted in a communication and maintenance overhead that impacted the delivery timeline. The timeline's significant impact can be attributed to two activities; designing the data model and generating data to validate the model. Reducing the time spent on these two activities meant more time spent on building the application's logic.
I first observed this fact as an internal consultant at a publicly-traded bank. I spent the majority of my time building an application that consolidated employee access to bank systems. We assembled a program that audited system access for branch and corporate employees. In this project, the data model's knowledge was distributed amongst external consultants, multiple engineering teams, and product experts. The following was common:
- Building data models from emails, spreadsheets, and meetings.
- Frequently updating the data design as new features came in.
- Waiting for data due to access delays resulting from regulatory guidelines, bureaucracy, or project dependencies.
I encountered this same issue as a solution architect at a consulting company. We were tasked with building a database application that determined insurance eligibility. The team consisted of project managers, data analysts, and database engineers on both sides. Once again, the knowledge of the data model was distributed amongst a large number of individuals. The following was also common throughout this project:
- Building data models from emails, spreadsheets, meetings, and slack chats!
- Waiting for data due to HIPAA (Health Information Privacy Act) processing requirements.
Again I encountered the same issues as a founding engineer at a data integration startup and as a tech lead at a publicly-traded online retail company. In both roles, we were tasked with building a data platform that consolidated data from various sources to support a single query environment. We commonly experienced the following:
- Product managers were consistently updating the data model based on new data sources.
- Sales members or project leaders were asking for contextualized demo data.
- Sister teams asking for sample data in different formats to build downstream systems.
- Stress testing systems with large volumes of simulation data.
In all these projects, a typical pattern emerged. Maintaining a data model and generating realistic simulation data helps teams overcome data access issues, build accurate prototypes and enable a transparent data environment. I realized these problems were commonplace in many organizations and groups. Furthermore, these activities were never given the required attention because they ultimately distracted the team from building the application at hand.
I searched for a single tool that would fulfill the following requirements:
- Manage a collaborative data design process between product managers, engineers, analysts and, other stakeholders.
- Generate simulation data from these data designs.
- Connect these data to an application, database, and/or analysis.
I never found a tool that tackled data design and data simulation broadly. Many applications such as database IDE's, API Mocking tools, and language libraries focus on their specific domain or technology.
So I decided to build one.
Whether your team wants to build a prototype while they await data access, demo the application based on the audience, stress test a system, or just focus on the problem. Overseed wants to help your team with their data design and data generation. We are building a platform that facilitates collaborative data design and provides realistic data generators.
Do you need these tools?
Want to give us feedback?