Company

System:  Specific details are protected by NDA. However, this system - developed over a number of years - involved creation of a unique architecture utilising LLM technology in a novel way. 

Client: NDA Protected

Technologies: Llama3-70B (and other LLM and ML models), vLLM, Python, Transformers, Algorithms

Problem: NDA Protected - but relates to asset price prediction

Solution: The innovative approach involved making an analogy between the "fuzzy" system we needed to model, and certain more rigorously defined mathematical/physical systems. Specific details are NDA protected. For the implementation we took advantage of per-hour cloud GPU rental to perform the necessary heavy processing as cost-efficiently as possible. 

System: Stock price history pattern matching utility

Technologies: Python (asyncio matplotlib mplfinance pandas...)

Problem: Build a system which identifies stocks that are currently in the process of repeating previously observed price history patterns, based on low-high and high-low trend time intervals.

Solution: Our solution involves iterative analysis of a large large database of stocks, gathered from web sources. Price histories are divided into progressively smaller time period bins, with duration labelled the "granularity" of the iteration. The pattern of trends across bins is compared with recent price movements to identify matching segments. We then used a visualisation system based on Japanese "renko" diagrams to depict matches in an easy-to-digest format.

Note this is a bespoke system created upon request; we provided no advice on stock selection methods.