microsoft/PhiCookBook
Oct 2024 – Oct 2024
1 Month active
Languages Used
GradleJavaKotlinMarkdownShell
Technical Skills
Android DevelopmentJetpack ComposeLLM IntegrationMediaPipeTFLite
Kinfey
PROFILE
Kinfey
Developed an end-to-end Android demonstration app for the microsoft/PhiCookBook repository, showcasing on-device large language model inference using the Phi-3.5 tflite model. The project featured a chat interface built with Jetpack Compose, enabling private, offline LLM interactions with low latency. The implementation included model loading, response generation, and comprehensive documentation to guide users through converting and deploying the model on Android devices. Leveraging Kotlin and Gradle, the work provided a reusable sample for customers to evaluate on-device LLM capabilities in real-world scenarios, streamlining onboarding and supporting practical exploration of private, efficient language model integration on mobile platforms.
Overall Statistics
Feature vs Bugs
100%Features
Repository Contributions
1Total
Bugs
0
Commits
1
Features
1
Lines of code
1,781
Activity Months1
Your Network
4728 peopleSame Organization
@microsoft.com4720
GitOpsMember
Ananta GuptaMember
Abi GicicMember
Abigail HartmanMember
Abram SandersonMember
Adam EttenbergerMember
Alexandre GattikerMember
Ami HollanderMember
AndersMember
Shared Repositories
8
Work History
October 2024
1 Commits • 1 Features
Oct 1, 2024October 2024 focused on delivering an end-to-end mobile demonstration of on-device LLM capabilities for PhiCookBook. The highlights include a new Android sample app that integrates the Phi-3.5 tflite model, featuring a chat interface, model loading, and response generation, plus documentation on converting and deploying the model. This work provides customers with a private, low-latency LLM experience on-device and establishes a concrete starter for evaluating on-device inference in real-world mobile scenarios.
1 Commits • 1 Features
Oct 1, 2024October 2024 focused on delivering an end-to-end mobile demonstration of on-device LLM capabilities for PhiCookBook. The highlights include a new Android sample app that integrates the Phi-3.5 tflite model, featuring a chat interface, model loading, and response generation, plus documentation on converting and deploying the model. This work provides customers with a private, low-latency LLM experience on-device and establishes a concrete starter for evaluating on-device inference in real-world mobile scenarios.
October 2024
Activity
Loading activity data...
Quality Metrics
Correctness90.0%
Maintainability80.0%
Architecture80.0%
Performance80.0%
AI Usage20.0%
Skills & Technologies
Programming Languages
GradleJavaKotlinMarkdownShell
Technical Skills
Android DevelopmentJetpack ComposeLLM IntegrationMediaPipeTFLite
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline