WhyHow.AI Platform Beta Update — SDK for programmatic flows

WhyHow.AI’s Knowledge Graph Studio is a modular platform designed to work seamlessly with your existing pipelines, and to be equally useful for developers and non-technical domain experts building KGs for systems of all levels of complexity. In line with this, we are excited to announce our SDK release (SDK documentation can be found here)

WhyHow.AI’s platform consists of 3 modular parts that we’ve built from the ground-up:

• Structured & Unstructured Data Ingestion to Graph Creation
• Graph Visualization, Interaction and Management
• Graph Querying Engine

Although you can build a KG RAG system with WhyHow.AI’s platform from end to end, we know every LLM-driven system looks very different. We want developers to plug in any part of WhyHow’s platform within their own custom LLM-driven workflows. As a database agnostic platform, you can export the graphs created with WhyHow.AI to graph databases of your choice, and we are building features aimed to support your data where they currently sit.

With our SDK, you now have:

• Flexible Data Ingestion & Framework Integration: Plug your existing data pre-processing pipelines into our graph creation process or Graph Studio
• Flexible Querying Logic & Answer Construction: Build on top of our Graph Querying Engine to enhance and customize the information coming from the graph
• Flexible Multi Graph<>Agentic Workflows: Build multi-graph multi-agent workflows with modular WhyHow graphs

Flexible Data Ingestion & Framework integration

We believe that:

1. There are multiple types of file formats we need to accommodate, and there are multiple types of data preprocessing pipelines that people are building already. This means that while context-preserving data extraction is still a difficult problem, many developers are comfortable building their own data extraction pipelines that WhyHow wants to accommodate and plug in on top of.
2. With the rise of a range of open source libraries for graph creation, some teams are interested in using their own processes for triple creation and are interested in using WhyHow’s platform for graph orchestration and management. Our platform is intended to be framework agnostic, and to accommodate all types of triple/ graph creation processes through our endpoints to plug in on top of.

Flexible Querying Logic & Answer Construction

We believe that:

1. Querying logic should be flexible. Our Graph Querying Engine can switch between both Structured & Unstructured Querying Engine. With our Unstructured Querying Engine, we take in natural language queries, transform them into the relevant nodes, triples and chunks to extract, with the option to turn the graph outputs into a natural language answer With our Structured Querying Engine, you have a more granular extraction logic that by specifying the entities & relationships you want to extract, and return all the relevant target entities.
2. Answer construction may take into account specific prompts, models or function calls. In more complex RAG systems, the way developers take the information from documents or knowledge graph information, and then transform it into your system’s own format and tone into the final answer for your customer varies greatly and should be accommodated. By plugging into our SDK, your answer construction pipeline can utilize the triples, nodes and chunks coming out of our graphs.

Flexible Multi Graph<>Agentic Workflows

We believe that:

1. The future of workflows is not just agent<>human RAG systems, but also agent<>agent multi-agent workflows. While information returned to a human in a RAG system can have some level of variance since humans can appropriately evaluate a returned answer, information passed between agents need to be far more accurate given the ability for error variances to propagate throughout a system.
2. We have multiple design partners that want to represent complicated domains to solve a customer problem. For example, imagine if you have a construction compliance use-case, and 3 documents. Document A is about Electrical regulations. Document B is about Plumbing regulations. Document C is about Labour Law regulations. You can imagine that you want separate graphs for each, since trying to merge these separate domains creates more complication than is worthwhile.

You will still need a Beta Access Code to access the platform and its API keys. For our Medium subscribers — One-Time Use Access Codes (If they are not working, they have been claimed already):

2ae1e14f-2160–4ce6–8441-f74c25a32287
c1935e47–6be4–45ae-a9f3–97f0dfb74b0c
e0a7e730-d50f-49a5–8d51-da0a367c2e67
82f94aa6–82aa-4836-a051-cd96bec43313
42af0a66–575e-45ba-9a80–08a7e61ee632
26ac8671–479c-422f-b689–050ca589e4ea

WhyHow.AI’s Knowledge Graph Studio Platform (currently in Beta) is the easiest way to build Agentic & RAG-Native Knowledge Graphs, combining workflows from developers and non-technical domain experts.

If you’re thinking about, in the process of, or have already incorporated knowledge graphs in RAG for accuracy, memory and determinism, we’d love to chat at team@whyhow.ai, or follow our newsletter at WhyHow.AI. Join our discussions about rules, determinism and knowledge graphs in RAG on our Discord.