The startup operates a communication platform that utilizes supervised learning algorithms to train neural networks for analyzing unstructured market conversations in OTC capital markets. This technology enables investment banks to extract detailed insights on liquidity, enhancing their ability to manage complex interactions with asset managers.
Funding
$893.1K raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
EFIN+1Founders
Product
Problem
Investment banks often struggle to efficiently extract actionable intelligence from the vast amounts of unstructured conversation data generated in over-the-counter (OTC) capital markets. The manual analysis of these conversations is time-consuming, prone to errors, and fails to capture the nuances of market sentiment and liquidity dynamics. This leads to missed trading opportunities and suboptimal management of complex interactions with asset managers.
Solution
Sense Street provides a communication platform that leverages generative AI and supervised learning to analyze unstructured market conversations within OTC capital markets. The platform extracts high-quality structured information from unstructured chats, enabling investment banks to gain detailed insights into liquidity, customer sentiment, and potential trading opportunities. By fine-tuning small language models (LLMs) with adapters that target specific extraction use cases, Sense Street delivers high-fidelity semantic outputs, including complex entities, missed RFQs, and primary market orders. The platform's APIs facilitate easy integration into existing systems, automating sales-trader workflows and enhancing the ability to manage complex interactions with asset managers.
Target Audience
The primary target audience includes investment banks and capital market participants seeking to enhance analytics, streamline workflows, and increase observability in their trading operations.
Features
- Utilizes generative AI and supervised learning algorithms to analyze unstructured market conversations.
- Extracts structured information from unstructured chats, including complex entities, missed RFQs, and customer sentiment.
- Employs small LLMs fine-tuned with adapters for targeted extraction use cases in capital markets.
- Offers both batch and interactive APIs for seamless integration into existing systems.
- Supports a wide range of asset classes, including rates, credit, commodities, money markets, and securities lending.
- Provides analytics to identify missed RFQs, decode counterparty sentiment, and retrieve conversational context.
- Incorporates a human-in-the-loop architecture for continuous feedback and model quality assurance.
- Ensures data security with enterprise-grade security protocols, ISO27001, and SoC2 certifications.