Skip to content
View katalinawinemixer's full-sized avatar
☺️
☺️

Block or report katalinawinemixer

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
katalinawinemixer/README.md

Katalina Londono

I have received the call to study the ancient language of machines.

I recently graduated with a B.S. in Applied Health Sciences and am based in San Francisco. My background spans laboratory workflows, clinical trial coordination, and regulatory affairs.

I'm moving toward healthtech, especially wearable and medical device research. Alongside that, I'm learning Python and JavaScript/TypeScript and building practical tools around study workflows, evidence review, and human oversight.

Projects

Healthcare Agent Ops Lab

A review workspace for healthcare agent responses. Reviewers compare responses with source evidence, score them, tag failures, and choose whether to approve, revise, or escalate. Approval gates check scores, evidence, privacy, and required handoffs.

The cases are synthetic, and review records stay in browser storage.

Demo · Code · Case study

StudyChaser

A regulatory training tracker for clinical study teams. It tracks protocol and amendment training by study and staff member, shows overdue acknowledgments, and previews follow-up emails with filing notes.

The hosted demo uses synthetic data, blocks record changes, and does not send emails.

Demo · Code

Human-AI Design System

Reusable interface patterns for citations, uncertainty, feedback, response comparison, and human review. Includes Model Behavior Studio for reviewing candidate responses and TrialSense for clinical-trial diligence.

Feedback selections carry into review exports. The app is a frontend prototype with static scenarios and local UI state.

Demo · Code

Spotted in SF

A food-photo app that ranks likely San Francisco restaurants and cafes. It strips embedded image metadata before provider analysis and presents guesses with uncertainty. Replacing or removing a photo cancels the pending analysis.

Demo · Code · Product notes

How I work

I start with a workflow problem, define what the tool should do, and check its behavior against those requirements. I use AI coding tools as part of my learning and prototyping process. I own the requirements, product decisions, review criteria, and verification, while I keep building my coding fundamentals.

I'm interested in clinical study operations, research coordination, regulatory operations, and implementation work with healthtech, wearable, and medical device teams.

Elsewhere

Projects and writing · Software portfolio · LinkedIn

Popular repositories Loading

  1. katalina-site katalina-site Public

    Personal site for Katalina Londoño — clinical-trial ops + LatAm biotech essays. Next.js 16 + MDX + Tailwind v4.

    TypeScript

  2. studychaser studychaser Public

    A lightweight regulatory training follow-up tracker for clinical research teams

    JavaScript

  3. human-ai-design-system human-ai-design-system Public

    AI-native design system for source grounding, uncertainty, feedback, evals, and human review.

    TypeScript

  4. sf-food-guesser sf-food-guesser Public

    AI-assisted learning prototype for guessing SF food venues from photos, with privacy-aware uploads, tests, and deployment checks.

    JavaScript

  5. workshop workshop Public

    Forked from raindrop-ai/workshop

    Give your coding agent the power to write and run agent evals.

    TypeScript

  6. katalinawinemixer katalinawinemixer Public