Shipped
Animated Metahumans (NVIDIA Audio2Face)
Production-ready experiment proving two animation paths: live microphone speech-to-face and synchronized audio-file playback-to-face for cinematic and gameplay workflows.
Outcome: Delivered repeatable facial animation workflows that reduced manual lip-sync effort and enabled faster narrative iteration.
Unreal Engine 5MetaHumanNVIDIA Audio2FacePythonOmniverse
Planned
AWS RAG Tutor
Learning assistant that retrieves trusted context and responds with references and confidence signals.
Outcome: Reduced unsupported answers by introducing retrieval checks and response grounding.
Next.jsAWS LambdaAPI GatewayDynamoDBOpenSearch
Planned
Research Agent
Agent system for topic decomposition, tool execution, and evidence-based synthesis.
Outcome: Improved consistency in research output quality through structured planning and validation.
TypeScriptNode.jsAWS Step FunctionsS3
Planned
Local LLM Coding Assistant
Developer assistant focused on private codebases and policy-aware suggestions.
Outcome: Creates a path for AI-assisted development in restricted environments.
PythonLocal ModelsVS CodeVector Search
Planned
LLM Evaluation Lab
Evaluation framework that compares prompts, models, and retrieval configurations.
Outcome: Faster architecture decisions with objective quality and cost signals.
PythonPytestPandasAWS
In Progress
AI Infrastructure Patterns
A set of practical infrastructure patterns for shipping AI features quickly and safely.
Outcome: Reusable templates that reduce setup time for new AI initiatives.
TerraformAWSCloudFrontLambdaDynamoDB