Research
Notes from the OpenKuber research team: what drove specific Indian stock moves, what AI models did when they traded NSE stocks, and how the causal engine is built. Every note explains past events and none is a recommendation.
How to Build AI That Explains Itself: Causal Ontology in IROHA
Most financial AI tells you what happened. We built the machinery to tell you why, traced to a verifiable source, structured as a graph, explainable to anyone.
The Map Forward: Stateful Agents and Causal Architecture
Trading is fundamentally stateful. When your agent forgets why it bought a stock 30 minutes ago, you lose. Here is how we fixed it with dynamic memory organisation.
Why Anakin Failed Performing in Indian Markets
A technically sophisticated AI trading system met a market that runs on signals no model was trained to see. This is what 167,000 lines of live execution logs look like when the thesis is wrong.
The Great Data Divergence: Why India is the New Quant Frontier
When LLMs meet EDGAR, there is no edge left. In Indian markets, the alpha lies in parsing unstructured data, decoding mixed-language sentiment, and seeing through regulatory PDFs.
What AI Trading Behaviors Reveal About How LLMs Are Built
By placing different LLMs in identical trading scenarios, the experiment revealed something the AI companies rarely discuss.
