Page not found. Your pixels are in another canvas.
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
This is a page not in th emain menu
As a new PhD student, starting at a new university, in a new city, in a new country, I’ve found myself introducing my work to many new people lately. I have worked in and (plan to, remember, just started) study machine learning for social and emotional interactions in robots. I’ve experimented with many ways to introduce this topic: “emotional intelligence, but for robots,” “social robotics,” “machine learning for robotic emotions.” Try as I might, I always get some flavor of this response:
“Women are socialized to…”
I’ve had many friends basically say they don’t believe “women are raised to…” and use this skepticism to put the burden of gender inequality on women. Here are some concrete examples I can think of that support the idea that, currently, women are brought up in society to…
An honest, cheeky, unfiltered brain dump of why I’m ready to go back. Partially an exercise to aid writing my statement of purpose, maybe somebody else can solidify
Obligatory Spoiler Alert.
Just some great pants.
Dark skin and Scottish accents and societally engrained biases, oh my!
Being a woman is really, really weird. As, I assume, being a man can be really, really weird. But I put forth the idea that being a woman leads to a constant rollercoaster of crazy second-guessing and walking a stupidly fine line between “being assertive” and “playing the victim.”
There has been a lot of talk recently about Deep Learning. Namely, how its recent resurgence is kicking ass across the board on machine learning benchmarks left and right. At the 2010 (and 2012) ImageNet Large Scale Image Recognition Competition, a group from University of Toronto, led by Geoffrey Hinton, built a neural net which outperformed the competition by 10% reduction in classification error rate (misrecognizing an image, so saying a picture that is in fact of a magnifying glass is a photo of scissors). This essentially brought neural networks back into play for the field of computer vision, where it dominates what most of us consider to be the coolest latest technology. Facebook’s face recognition? Neural network. Google’s self-driving cars? Controlled by neural nets. The tempting logical jump to make here is: artificial intelligence? Let’s just use a neural net!
When loving, well-meaning men in my life don’t understand the problem with gender separation, I try to write.
Mentor and Organizer
- Instructor and curriculum designer for the Girls Computer Club for secondary school students.
- Recruited 20 girls from local secondary schools.
- Taught computer science basics, including sorting algorithms, pointers, and python basics.
- Coordinated logistics of club hackathons, speakers, and outreach.
- Led executive board in creating WiCS mission and club constitution.
- Grew club 4x over two years.
- Instructor for Girls Who Code program.
- Supplemented given curriculum with weekly lessons on computer science basics.
- Brought in speakers for work presentations, demos, and artists who employ technology in their work.
Volunteer Mentor & Organizer
- Organized and designed engineering curriculum for Cambridge Inventors Club.
- 40 children age 11-13 each term participated weekly in engineering challenges, including engineering design principles like iterative testing and collaborative working.
Mentor and Organizer
- Cogito corporate sponsor and technical mentor for Resilient Coders -- an organization which trains non-traditional background adult students to become software developers.
- Developed a 8-week "boot camp" curriculum for two apprenticeships.
- Trained apprentices in software development practices, and in presenting to company.
- Troupe volunteer for Girl Guiding program, ages 9-11.
- Volunteer weekly with troupe leading leadership and cooperation building skills.
- Supervise weekend trips to museums and venues around Glasgow.
SRC Class Representative
- Gathered feedback on courses, program structure, and student experience.
- Participated in Student Council meetings to represent program cohort.
- Collaborated with faculty and staff on program and messaging improvements.
- Volunteer for STEM Ambassador program.
- Participate in career talks and robotics demos for high schoolers.
- Facilitate and mentor high school robotics team.
- On Student committee for department Athena SWAN chapter.
- Designed and implemented initiatives to encourage students to become STEM Ambassadors.
- Implemented metric gathering mechanism to track demographics of seminar and workshop participation.
- Collaborated with sociology department on evaluation metrics and surveys to ensure validity of efficacy reports.
Tufts University, Department of Biomedical Engineering
- Research Assistant for Professor David Kaplan in silk lab.
- Developed directional freezing technique for silk solution.
- Trained in high-powered electron microscopy.
Tufts University, Department of Computer Science
- TA for CS courses: Introductory, Data Structures, Assembly Language Programming, and Machine Learning.
- Ran labs, graded assignments, held open office hours and 1-1 tutoring sessions.
- Participated in redesigning course curriculum for Machine Architecture and Assembly Language Programming.
- Full stack feature development of internal desktop application.
- Developed web APIs and GUI for music annotation tools.
- Employee 20(/110)
- Designed a series of tutorials from ”Getting Started” to advanced usage to onboard internal and external developers.
- Architected data APIs for dev tools to manage robot sensory input from cameras and microphones.
- Implemented frontend UI (React) and developer interfaces to build high-level skills.
- Won inaugural company culture award for "outstanding cultural contributions."
- Designed and implemented robotic bi-axial Emotion System and corresponding impacts on robot behavior.
- Fully implemented and maintained robot listening pipeline (main interface for interaction).
- Full-stack development of Role-Based User Management (RBUM) APIs
- Front-end React feature development
- Continuing work to rearchitect APIs in cloud-based architecture.
- Merged two software teams, technically and personally managed team of 10 full-time engineers.
- Oversaw top-to-bottom development and implementation of three client-facing features.
- Hired and onboarded three product managers, including one for another software team.
- Acted as Scrum master leading agile rituals (standup, planning, retro).
- Implemented and managed release pipeline, and later trained release manager.
Bi-Axial Emotional System
Genetic algorithms applied to LED badges
Engineering for Kids!
Technical Workshop and Judge
Technical Workshop on Jibo API: Quickstart
- Getting Started
- Making Animations
- Controlling Application Flow
- Taking Advantage of Native Infrastructure
- Connecting to the Outside World
- Testing Your App
It’s 2016, and humans don’t give instructions through terminals anymore. The future of interfaces is social. Natural language understanding, conversational dialogs, and subtle social cues are the new instruction set – what challenges exist in the vast field of HCI, and how can current machine learning techniques address them?
"Alexa, Read the Room."
Emotions govern so much of human behavior – why are voice agents taking so long to catch up? What does the future of an emotional voice interface look like, and how will the ability to perceive and express emotions influence the development of voice interfaces in the future?
Social Robotics Workshop
What is a social robot? What applications are social robots uniquely suited for, and how can we design interactions that are useful, and socially fulfilling?
There’s earnest self-improvement and criticism, and then there’s crippling self-doubt. Learn techniques to differentiate between subtlties of the two, and identify self-sabotaging behaviors you can mitigate to reach your highest potential.
Abstraction and API Design
How do you design an API for a programmer? Too little control, and your audience is frustrated by a lack of power. Too much detail required, and your developers can no longer keep track of everything to specify. How do you strike a balance between power and specificity to write the most useful library?