Colin Banigan

Full-stack software engineer on the New Bets team focused on creating new customer experiences that extend Instacart’s growth opportunities. Highlights:

• Current: Tech Lead for Instacart’s AI strategy with external third-party chatbot providers. Defining AI-native commerce experiences (MCP, OpenAI Apps SDK, etc.) with frontier AI labs to enable grocery and essentials shopping directly through chat, while building the underlying systems that make the experience feel seamless and magical.
• Current: Tech Lead for the First Order Experience team, optimizing the conversion funnel from app start to first order placed. Partner with engineering leadership and product to define the roadmap, review engineering approaches, analyze experiments, and mentor engineers across the team.
• Tech Lead for the Weekly Flyers experience in the Instacart app. Led a cross-org engineering effort spanning data acquisition pipelines, catalog storage, API orchestration and serving, client experience, and CRM email and push engagement. Drove alignment across engineering, product, data science, design, BD, and marketing to launch a comprehensive savings feature used across the app. Launched flyers on the majority of Instacart’s primary retailers and delivered multiple $ savings per order with flyer items.
• Founding engineer (1 of 4) on the Restaurants initiative, expanding Instacart’s capabilities beyond groceries to include restaurant ordering. Rapidly prototyped an in-app restaurant experience to demonstrate the vision and help solidify the partnership between Fidji Simo (Instacart) and Dara Khosrowshahi (Uber).
• Founding engineer (1 of 3) on the Fizz team, a startup-style initiative within Instacart focused on party order delivery. Built a new app and brand powered by Instacart infrastructure to reach a demographic that historically did not use Instacart. Represented engineering in the Partiful integration and led authentication, retailer selection, CRM, and post-checkout experiences.

Full-stack software engineer on the Growth Activation and Acquisition teams focused on converting visitor traffic into paying customers. Highlights:

• Authored Instacart’s internal Rich Text framework, enabling developers to define server-side styling for rendered text across Web, iOS, and Android. Identified both a user need from retailer and marketing tooling requests and a technical gap where engineers were re-implementing styles across the three clients. Delivered foundational extensions to the internal API framework, and the resulting system is now used by all teams in the Core Experience organization with hundreds of distinct usages.
• Led development of the Self-Serve Landing Page System. Observed that engineers across the organization required roughly two weeks to build new landing pages end-to-end and standardized the landing page component architecture in partnership with the Instacart Design System team, reducing it to nine core components. Defined server–client contracts across API and storage layers, and built custom CMS page builder tooling to support no-code creation and visualization. Reduced page implementation time from two weeks to two hours, enabled PMs and Marketers to be the primary page creators, and have launched hundreds of distinct landing pages to date.

Full-stack software engineer on the Growth Acquisition team focused on new customer acquisition. Highlights:

• Primary engineering owner of the Instacart Homepage and Product Landing Pages. Led the re-architecture effort with a team of 6 engineers from Ruby on Rails with client-side rendering to a Node-based server-side rendering system, enabling parallel API calls via GraphQL and improving overall performance and reliability. Cut page serving p90 from ~3s to ~1s alongside A/B experimented growth feature enhancements.
• Drove Instacart shopper landing page growth during early 2020 when customer demand far exceeded shopper availability. Created the “jobs near me” SEO landing page ecosystem, introducing a new gig-economy SEO strategy centered on long-tail, location-based job keywords. This work supported the onboarding of roughly 500k shoppers in March and April 2020, providing needed work opportunities during a challenging economic period.

Relevant technologies: Ruby (Ruby on Rails, Rspec), JavaScript/TypeScript (React, Node.js, Express, Jest), Python (Apache Airflow, FastAPI, pytest), Golang, Amazon Web Services (AWS), Segment, Google Tag Manager (GTM), DataDog, Sentry, SQL (PostgreSQL), NoSQL (DynamoDB), HAML, SASS, HTML, CSS

Software Engineer within Data Engineering in Capital One Financial Services. Re-engineered the data pipeline from on-premise to the cloud (AWS). Work included distributed computing for data load, transformation, and validation (PySpark + AWS EMR + S3 + Snowflake), full-stack web development (Angular + Java Spring Boot/Python Flask + PostgreSQL), autonomous deployment on an internal CI/CD pipeline with unit, integration, and performance tests (Docker + Jenkins + Python Unittest + Behave + JMeter), and machine learning data analysis (Python + H2O).

Full-stack developer within the Home Loans Department. Fully integrated a core feature in the main Home Loans application to vizualize and sort previous contact information to improve customer satisfaction. Project included creating a UI in Angular 4, writing an OL in STS, and retrieving data from Cassandra DBs.

Android developer and researcher at The StoryLab. Primary research focused on informal science education through the use of Android Wear smart watches. Secondary research compared and contrasted movement within virtual reality.

GPA — 3.8
Master of Science in Computer Science with a specialization in Interactive Intelligence from Georgia Institute of Technology.

Completed two graduate classes while pursuing my undergraduate degree at Texas A&M:

• CSCE 671: Computer-Human Interaction
• CSCE 625: Artificial Intelligence

Completed eight graduate classes with Georgia Tech:

• CS 7637: Knowledge-Based AI
• CS 6300: Software Development Process
• CS 7646: Machine Learning for Trading
• CS 6440: Intro to Health Informatics
• CS 6035: Introduction to Information Security
• CS 7638: Robotics: AI Techniques
• CS 8803: AI, Ethics, and Society
• CS 7641: Machine Learning

GPA — 3.57
Bachelor of Science in Computer Science from Texas A&M's Department of Engineering. Completed two minors in Visualization and Cybersecurity, and a Business Management Certificate with the Mays Business School at Texas A&M. Graduated Cum Laude.

Abstract:
Relatively little research exists on the use of smartwatches to support learning. This paper presents an approach for commodity smartwatches as a tool for situated reflection in elementary school science. The approach was embodied in a smartwatch app called ScienceStories that allows students to voice record reflections about science concepts anytime, anywhere. We conducted a study with 18 fifth-grade children to investigate first, the effects of ScienceStories on students' science self-efficacy, and second the effects of different motivational structures (gamification, narrative-based, hybrid) designed into the smartwatch app on students' quality and quantity of use. Quantitative results showed ScienceStories increased science self-efficacy especially with a motivational structure. The gamified version had the highest quantity of use, while narrative performance performed worst. Qualitative findings described how students' recordings related to science topics and were contextualized. We discuss how our findings contribute to understanding of how to design smartwatch apps for educational purposes.

Abstract:
This paper extends the concept of digital literacy and applies it to Making. Through case descriptions, we contribute an understanding of how children can become or fail to become individuals literate in Making within a formal learning context. Our analysis draws from video recordings and other data sources of two 4th grade classrooms in which the students, who had already participated in 1.5 years of more structured 'makified activities', engaged in an open-ended, exploration-based, and playful task that was more in line with the spirit of Making. Student teams were classified as 'high in Making literacy' and 'low in Making literacy', revealing how Making literacy was expressed at the level of skills, mental models, and practices in various ways for different students. Our qualitative analysis demonstrates what burgeoning Making literacy may mean in a public elementary school classroom, paving the way for a vision of a time when Making becomes generalized practice.