📈 Data Platform Priorities - Large Language Model - LLM (Epic)

[Ed-Bajo]

Brief summary of what this Epic is

User Story

As a Data Scientist
I want to explore using Large Language Model (LLM),
So that there is a secure managed location to explore using these tools that can aid MoJ's data driven strategy.

Other information

The Analytical Platform currently does not have the ability to offer LLM tooling due to the limitations.

Discussion on LLM in ask channel here

Value

• provide tooling to aid our Data Science teams with technology that can speed up delivering value from our data
• reduction of repetitive manual tasks, Improving communication, entity extraction, text summaries, voice and image recognition.
• reduce teams maintenance and management by using a Managed Service offering

Questions / Assumptions / Hypothesis

Hypothesis

If we provide tooling to our data scientists
Then we can tell better stories about our data

Proposal

Carry out a POC on integrating Bedrock onto the Analytical Platform.

Assess the capability, operational requirements and costs of Bedrock and compare it to other tools

Definition of done

• Integrate Bedrock (LLM) tool on to the Analytical Platform.
• Identify users for the LLM tool and gauge the performance
• Put in place mechanism for capturing user feedback
• Consider how to integrate the identity management and access control into the tool
• Ascertain cost analysis
• POC findings documented

List of issues in this epic

• 👥 Identity and Access Management #1132
• 🔒 Authorisation microservice #2078

Other information

No response

• Spike: Breaking Down the LLM Epic #3000
• 🤖 Identify an LLM Quick Win #3456
• 🛩️ Refactor iam_builder for terraform #4290
• ☁️ Complete Deployment of Bedrock into Airflow #4718

[github-actions]

This issue is being marked as stale because it has been open for 60 days with no activity. Remove stale label or comment to keep the issue open.