SIGnopsis: Edge AI R&D PI -

Role: Chief Architect - FY24 Accenture Oracle Mastermind Hackathon

This tool demonstrates how to use AI at the edge for raw data to executive synopsis & intelligence reporting for agnostic raw source data via a land-air-water-space network data fabric mesh.

How It Works

• The core is a Go program named "synopsis.go".
• It frontloads AI API to generate executive synopsis and intelligence reporting that are relevant to raw data context at the edge.
• A frontend (e.g., built with a template like the UFO Alien Template) provides a user interface for interacting with the chatbot.
• This Proof of Concept can be leveraged on Cohere's Powerful command-r-plus model at 128k token output on AuthN/AuthZ silo'd LOB data

File Structure

edge-ai-service/
├── .models.{theoretical}/
│   └── x.ai.grok.alpha.200b.params
│   └── mistral.latest.bin
│   └── llama.latest.bin
│   └── custom.bin.public.private.access
├── pkg/
│   └── synopsis/service.go
|   └── shared/subjects.go
├── cmd/microlith/
│       └── synopsis.go
├── .env
├── Dockerfile
└── README.md

Architecture

Refer to Pretty one at ./artifacts/architecture.png

graph LR
A[Edge Services Admin] --> B[Corsight]
B --> C{Python 37}
C --> D[OCI EDGE]
C --> E[Redis] 
C --> F[Software]
D --> G[OCI Admin]
D --> H[Application Admin]
D --> I[Videos Stream / Snapshot Push]
D --> J[Image or Video Capture]
I --> K[ULTRA]
G --> L[OCI Chicago]
G --> M[OCI Ashburn]
L --> N[OSN]
N --> O[Object Storage]
N --> P[Autonomous Database]
N --> Q{ULTRAMQ AP}
Q --> R[AI Vision]
Q --> U[Logging]
Q --> S[APEX]
Q --> T[Auditing]
M --> V[OSN]
V --> W[Object Storage]
V --> X[Autonomous Database]
V --> Y{ULTRANATSMQ AP}
Y --> Z[AI Vision]
Y --> AD[Logging]
Y --> AB[APEX]
Y --> AC[Auditing]
H --> AE[ERP User]
H --> AF[SCP User]
H --> AG[Vegetation Client App]
H --> AH[LOB Analyst]
H --> AI[LOB Analyst]
L --> AJ[XYZ Region]
AJ --> AK[Synadia Cloud]
AK --> AL{Cohere}
AK --> AM(OpenAl)
AK --> AN[[AWS]]
AK --> AO[[Azure]]
AH --> AP[3rd Party Interface]

Loading

Requirements

• A Cohere API key (https://cohere.ai)
• Go programming language (https://go.dev/)
• A frontend web framework or HTML template (optional, for the user interface)
• Nats.io Edge Messaging Fabric Technology (https://nats.io)

Installation

1. Obtain a Cohere API key: Sign up for a Cohere account and get your API key.
2. Install Go: Follow the instructions at https://go.dev/doc/install
3. Clone or download this project: This will give you the synopsis.go file.
4. Set Environment Variable: Set the CO_API_KEY environment variable with your Cohere API key.
   • Linux/macOS: export CO_API_KEY=your_api_key; export NATS_URL=your_nats_global_super_cluster
   • Windows: Use the System Properties settings.

Development and Testing

# Add your nkey credentials file
-----BEGIN NATS USER JWT-----
uuidv40xDAGeysuperdupersecretencryption123ociwoohoosafracupnerd1241235123451235rqegrq245h
------END NATS USER JWT------

************************* IMPORTANT *************************
NKEY Seed printed below can be used to sign and prove identity.
NKEYs are sensitive and should be treated as secrets.

-----BEGIN USER NKEY SEED-----
superdupersecretencryption123ociwoohoosafracupnerd1241235123451235rqegrq245h
------END USER NKEY SEED------

*************************************************************

# Subscribe to synopsis messages anywhere🤓
nats sub "edge.ai.synopsis"

go mod download
go run ./cmd/microlith/synopsis.go

Building and Running

1. Build the Go binary:

go build ./cmd/microlith/synopsis.go

This will create an executable file named synopsis (or synopsis.exe on Windows)

2. Run the chatbot:

./synopsis -i "<co: 0>341 RSA2048-10-11T22:14:15.003Z edge.machine.oci-rover.ai su - ID47 - BOM'su root' failed for lonvick on /dev/pts/8"

Use code with caution. The chatbot will start listening for input from the command line.

Frontend Integration (Optional)

Choose a frontend web framework (e.g., React, Vue.js, Svelte) or use a pre-built template like the UFO Alien Template.

Implement logic in your frontend to:

Send user queries to the synopsis executable (potentially running on a server). Display the chatbot's responses from Cohere within the user interface.