top of page
Workflow Builder

At Strateos, our mission was to turn the life sciences into an information technology, driven by data, computation, and high-throughput robotics with the goal of substantially advancing drug discovery. Our team aims to bring together deep expertise in biology, automation, and software engineering to set new standards for the way scientific research is done.

Why workflow Builder?

Strateos clients, whether they be hybrid cloud lab or remote assay execution clients, need a way to represent experiments they conduct on our platform. Currently, this representation is rate-limited by the number of engineers we have available to transform the client’s English protocol into validated (both virtually and biologically) Autoprotocol-Python.

 

We propose  an infrastructure which would allow scientists access to our platform, execute experiments capturing scientific intent and complete workflows, without the need for a custom implementation and/or human support to create the workflow.

Workflow is a custom blueprint that users design to create and submit experiments.

Experiment is each submitted workflow

Problem

Onboarding potential clients to Strateos platform requires redundant internal work and wastes time.

Sunk Costs on Lost Leads

  • Long onboarding times directly tied to lost leads.

  • Invested time in lost client cannot be recouped.

 

Siloed Feature Development

  • Features developed for one client cannot easily be made available for another.

  • Product offerings do not expand with clients served.

 

Cognitive Burnout

  • Protocol scripting is repetitive and does not scale.

  • One application scientist is tied to one client.

  • Causes burnout and institutional knowledge loss.

Screenshot 2023-04-18 at 12.08.14 PM.png
Process

This would be a long project, so I defined a process and timeline to keep myself on track :) 

  • User Research: Develop my understanding of auto protocols and experimentation domain through in-depth interviews with potential users and subject matter experts.

  • Design Sprint: Lead a 1 week Design Sprint to bring together all stakeholders, executives and subject matter experts to co-create an initial prototype.

  • Synthesis: Drawing together all material from user research and the Design Sprint, synthesize findings into personas, journey maps, requirements jobs-to-be-done and user acceptance tests.

  • Solutioning: Drawing on the synthesized research material, iterate on UI mockups, performing user testing after every iteration.

User research / Discovery

Tag teaming with the PM, i kickstarted user research to understand core user needs and interactions we need to define for workflow builder. I typically use tools like user studies , ui flows and journey maps to capture and synthesize user touch points , but before we get started with any tools we needed to define base understanding of the problem and solution area.

Screenshot 2023-04-18 at 12.16.17 PM.png

Workflow for protocol implementation and testing

Interviews

I conducted in-depth interviews with approximately 20 anticipated users of the Strateos chemistry platform — all from various pharmaceutical companies. I conducted my interviews on-site at campuses in San Francisco, Menlo park, San Diego and leverages Usertesting.com to source external users.

 

The roles of these users included Auto protocol writers, Quantitative Biologists, Assay Development Specialists, Medicinal Chemists, Computational Chemists, and thought leaders in the realm of building and executing experimental workflows.

Synthesis

My interviews and shadowing sessions lasted about 1 and 2 hours each and I recorded and transcribed them. I then reviewed and annotated each transcription to break out common themes such as concerns, pain points, ambitions and core requirements. These syntheses were invaluable later on as I began to prototype solutions.

Accessing the Strateos cloud lab is impossible without institutional knowledge and programming skills.We need to enable internal and external users to parametrize and submit experiments to Strateos to truly provide open access science.


 

  • Accessibility: Anyone should be able to submit an experiment to Strateos.

  • Reusability: Any experiment executed at Strateos, regardless of user, organization, etc, should be reusable in future experiments.

  • Scalability: Any new experimental functionality should be able to be encoded and deployed to this service, without custom code needing to be written for a client

Screenshot 2023-05-01 at 11.46.02 AM.png

Outline of user needs, possible solutions, and company goals these solutions align with.

Design research and testing

I conducted market research to review and analyze existing workflow builder patterns and wanted to validate some of my assumptions on how our scientists would like to interact with the workflow builder.

After several meetings with stakeholders and user interviews , i wanted to quickly test with simple Miro prototype to gage some of the interaction pattern 
and found out most of our users would like to build workflows in a horizontal way and needed more data input upfront

Screenshot 2023-05-01 at 11.56.15 AM.png
Screenshot 2023-05-01 at 11.56.23 AM.png
Optimization

During the optimization process, we’ll slowly trim away the unwanted fat and work with our cross functioning teams to identify the core needs and workflows. 

Here you can see an example of a early UI flow we were mapping out possible entry and exit points for the feature

Frame 1.png
Solutioning

Finally after weeks of research and planning, it was time to dive into prototyping. I would draw on the additional synthesis and scoping done since the Design Sprint to build on the prototypes developed during the Design Sprint. Over the course of eight iterations, I developed a full flow of the proposed user experience.

RAPID ITERATION

To begin, I started with rudimentary sketches to quickly iterate on basic concepts. Shown here is a small sampling of these first iterations. I matched these concepts against the Acceptance Tests to ensure they addressed the identified requirements. I also validated them with a number of various user types to ensure the ideas and concepts were sound.

These preliminary sketches served as the source material for a second round of prototyping which detailed a proposed solution in full, screen by screen

Screenshot 2023-05-01 at 2.22.49 PM.png

LOFI Sketches

Screenshot 2023-05-01 at 2.23.07 PM.png

HIFI mocks

Visual Design

At this phase of our product, the visual design for the mocks making up this flow was only partially refined. I was focusing on the user flow and information hierarchy/layout. Some screen shots for reference :)

bottom of page