Its ability to connect to a specific company knowledge database.

I really appreciate the integration of local and global intellectual capital repositories within the library, together with the solution design workflow embedded into the user interface via Projects, and the capability to expand teamwork both in content creation (Project) and in content review (Porject). AI specialization from analysis to review is helpful, supported by some custom prompting to generate storyboard parts. I've also interacted with AutogenAI Support Team, issuing some tickets, and I promptly received feedback and resolution. I've experienced a performance boost in my traditional scavenging approach to find good intellectual capital reusable content thanks to Research, and I appreciated the Ideator feature. I used a lot the Editor We Will, Explain How, Add Case Study and Add Statistics functions, powered by company's repos are strong improver of generated content. The capability to sub-set the IC repos by Source Provider feature is also great, because you can taylor your content according to market segment, year, or specific artifact type, at least if the IC repos are well setup. For example: I use Extract to collect info from specific topics that could be spread into several requirement documents I use Ideator to craft ideas turning on all the AI embedded in AutogenAI; useful ideas are converted into Drafts documents, to be used as valuable insert into a proposal document If some diagrams are included in the document set, I analyse them using Copilot or ChatGPT, obtaining a text description of the diagram, being an org model, a high-level plan, an AOD, or similar. This enable AutogenAI to enrich its AI Analysis using content it can actually process I use Search function within AutogenAI to look for interesting artifacts or IC that could be reused, when targeted by Sources selection within AutogenAI I craft a prompt in Research to build prompts for REQUIREMENTS section in storyboard that are not autogenerated by AI Analysis in Project, due to custom build of proposal outline I draft the Q&A questions list to be submitted to the Customer, together with engagement team, and generate an answer using Workbooks and Library AI; this is not intended to be meaningful to some questions, but it has proven to be informative before actual Customer answers. I build the proposal draft within AutogenAI supported by fellow solutioners, using their knowledge on the topic to enrich the drafting with meaningful prompts; currently AutogenAI doesn’t provide followup prompts nor in Research nor in Editor, while Copilot or ChatGPT do. I use only Library AI and Datasets, narrowed to market sector or specific topics if I know these are available in KL I use Gamma Review to suggest compliance checks with requirements, and feasible suggestions are used within AskAI as prompts After building a proposal draft within AutogenAI, I export it to be uploaded and shared into the engagement folder, allowing other engagement people to explore its content, and use it to craft their proposal sections content; if available from other solutioners, I add images / diagrams / tables

Its giving good answers spanning across multiple documents. its quick as well