DT Global

DTLLM and Evidence Portal

Structured evidence access and a reusable portal for internal learning and project workflows.

Project overview

What this project was built to help with

DT Global's evidence base spans internal records, academic research, and publicly available competitor evaluations. DTLLM brings that material into a custom large language model and evidence portal, allowing teams to search, analyze, benchmark, chat with evidence, and generate synthesis for internal learning, bid strategy, and project workflows.

Project story

Unlocking DT Global's institutional evidence base

DT Global needed faster access to insights across internal records, academic research, and publicly available competitor evaluations. The evidence base was large enough that normal search and manual synthesis could slow down learning, bid preparation, and project-facing work.

DTLLM combines a custom large language model with an evidence portal. Users can search, analyze, benchmark, chat with evidence, and generate synthesis or document drafts. The tool helps teams understand performance, strengthen internal learning, improve competitive positioning, and spend more time on strategic thinking rather than evidence retrieval.

Example deliverables

Concrete outputs or working assets the project produced or supported.

Deliverable 01Custom large language model for DT Global evidence
Deliverable 02Evidence portal combining internal data, academic research, and competitor evaluations
Deliverable 03Search, chat, benchmarking, and synthesis workflows
Deliverable 04Automated document creation support for learning and bids
Development team reviewing evidence together in a meeting

Decision problem

DT Global needed a way to unlock real-time insight from a large internal and external evidence base, including internal records, academic research, and publicly available competitor evaluations. Teams needed search, synthesis, benchmarking, and chat-based evidence interaction for learning, bid strategy, and project-facing workflows.

What the work involved

1

Built DTLLM as a custom large language model and evidence portal for searching, analyzing, and benchmarking DT Global's evidence base.

2

Combined internal data with academic research and publicly available competitor evaluations so users could compare evidence beyond a single internal archive.

3

Enabled chat-based interaction with evidence and automated document creation, helping teams move faster while focusing staff time on strategic interpretation and bid positioning.

Outcome

More consistent access to project knowledge and evidence for delivery teams.

Evidence inputs

What the workflow brings together

Project recordsEvidence productsInternal learning assetsProgram knowledge

Review focus

What stays visible for human review

Search quality
Evidence interpretation
Reuse across teams

Talk with us about a similar workflow.

Bring one decision, report, review process, or evidence problem. We will help identify the smallest useful place to begin.

Start the conversationExplore AidInsight
Cookies

We use cookies to enhance your browsing experience on our website. By continuing to use our site, you consent to our use of cookies. For more information, please review our Privacy Policy.