Justin Sepulveda

Microsoft

2022 – Present

I'm leading design in the rapidly evolving space of organizational analytics, where my team ships tools that help companies understand their workforce and AI investment impact.

• Agent Dashboard We designed a dashboard that shows companies how their AI agents are being used and where there's opportunity for greater impact.
• Copilot Dashboard We created the analytics experience that helps companies understand Copilot's impact and how to maximize its value.
• Skills My team designed tools that help companies see what skills their people have and where to invest in development.
• Pulse We shipped a feedback tool that gives managers real-time insights from their teams.
• Organizational Data Ingestion System We designed the system that lets companies safely connect their organizational data across Microsoft's tools.

Amazon CodeCatalyst

2021 – 2022

I led the design team that built CodeCatalyst, a unified dev environment that brought planning, coding, and deployment together instead of forcing developers to context-switch between tools.

Shipped: Amazon CodeCatalyst

Amazon Shopping

2019 – 2021

I led a design team that improved product detail pages for hundreds of millions of Amazon shoppers while navigating a complex stakeholder landscape across Amazon.

Shipped: All Offers Display, Multi-Offer Display, Pricing Display, Structured Attributes, Twister+

AWS Machine Learning and Edge Computing

2016 – 2019

I built AWS's ML design practice from scratch in the early days of machine learning, leading teams that launched 20+ products across computer vision, NLP, human-in-the-loop systems, and edge computing.

Machine Learning

• Amazon A2i Human review service for machine learning predictions, enabling organizations to integrate human judgment into ML workflows.
• Amazon CodeGuru ML-powered code reviewer that identifies performance issues and provides recommendations to improve code quality and reduce costs.
• Amazon Comprehend Natural language processing service that extracts insights like sentiment, entities, and topics from text without ML expertise.
• Amazon Comprehend Medical Medical NLP service that extracts clinical information from unstructured medical text with high accuracy.
• Amazon Forecast Time-series forecasting service that predicts business outcomes like demand and staffing needs using the same ML technology as Amazon.com.
• Amazon Kendra ML-powered enterprise search service that delivers accurate answers from multiple content repositories using natural language queries.
• Amazon Personalize Real-time personalization and recommendation service that delivers hyper-personalized user experiences across digital channels.
• Amazon Polly Text-to-speech service that turns text into lifelike speech in dozens of languages with multiple natural-sounding voices.
• Amazon Rekognition Computer vision service providing pre-built and customizable APIs for analyzing images, videos, and video streams at scale.
• Amazon Textract Document analysis service that automatically extracts text, handwriting, and structured data from scanned documents using ML.
• Amazon Transcribe Automatic speech recognition service that converts audio to text, supporting 100+ languages with speaker identification and content redaction.
• Amazon Transcribe Medical Medical speech recognition service designed for clinical documentation with medical terminology and specialty-specific accuracy.
• Amazon Translate Neural machine translation service for fast, customizable language translation to localize content for global audiences.
• AWS DeepComposer Machine learning-enabled musical keyboard that teaches generative AI by composing original music.
• AWS DeepLens Deep learning-enabled video camera for developers to build and deploy computer vision applications.
• AWS DeepRacer Autonomous 1/18th scale race car designed to test reinforcement learning models on physical track.
• AWS SageMaker Ground Truth Data labeling service with expert support for fine-tuning models to achieve specific accuracy and performance goals.

Edge Computing

• AWS Cloud Map Cloud resource discovery service that maintains a dynamic registry of microservices, databases, and queues with automated health monitoring.
• AWS Elemental Media Services Comprehensive portfolio of purpose-built media services that process, transport, encode, package, and monetize video content at global scale through cloud-based solutions.
• AWS Global Accelerator Networking service that improves application availability and performance using AWS global network infrastructure.
• AWS WAF Web application firewall that protects applications from common web exploits and bots that affect availability and security.

AWS

2014 – 2016

I designed for AWS database, analytics, and AI services during the early days of cloud-native development.

• Amazon DynamoDB Serverless NoSQL database with single-digit millisecond performance at any scale, automatically handling infrastructure management.
• Amazon Lex Conversational AI service for building voice and text chatbots that understand intent and maintain context across multiple languages.
• Database Migration Service Managed migration service for securely moving databases to AWS with minimal downtime and zero data loss.

Thermo Fisher Scientific

2006 – 2014

I built research tools for life sciences, helping scientists design custom genetic experiments and order materials.

Shipped: Custom Array Designer, Custom Assay Search, GeneArt Gene Synthesis Tools, Life Technologies Design System