About Us
The Client Problem
 
  • Massive amounts of data on the web being produced every day.
  • Decision making is increasingly automated for many businesses.
  • People don't have time to read or interpret thousands of words per second.
  • Actions often need to be taken within seconds or minutes.
 
The TOPOS®  Solution
 
  • Proprietary methods for refining high volume, high velocity real-time data into numbers that can be used for decision making purposes
  • A subscription model that allows clients to discern refined data obtained from Twitter's live or historical data via APIs and/or a configurable UX.
 
Competitive advantage
 
Cost: TOPOS® offers sentiment  indicators  at  a  much  lower  cost  than  would  be possible if  the  client  were  to  develop  such  indicators  independently.  Access to the Twitter feed   will be discounted by more than 70%.  This is possible because TOPOS® has a scalable product.    Of course, TOPOS can't simply re-syndicate the Twitter feed to clients, but our hypothesis is that the true value in this data is in what can be refined from the raw text.
  
Accuracy:   Our base indicators have accuracy levels that are remarkable.   Our advanced Deep Learning methods are even better.  But on top of that, with our patent pending method for classifying Tweets we believe we will be in a position to have the most accurate Tweet Sentiment Indicators.
 
Low  Latency:  Using  methods  derived  from  high  frequency  trading,   TOPOS® will deliver updates  to sentiment  indicators  at  the  lowest  latency  possible.  
      
Flexibility: TOPOS® will offer tailor-made services to clients who wish to focus on specific segments or parameters.
 
Management Team
 

                                             




                                 

Marc da Costa Nunes holds a Ph.D. in Mathematics from the University of California, Berkeley                   and a B.A. (with Highest Honours) in Mathematics from the University of California at Santa Cruz. Marc has over twenty   years of experience in quantitative analysis, statistics, software development, risk management and capital markets. He has also built high frequency statistical arbitrage models for Société Générale’s High Frequency Trading Desk.  For ten of his seventeen years at Société Générale, Marc was a senior Managing Director in charge of Global Hedge Fund Risk.  He succeeded in developing an award-winning quantitative model to detect fraud in hedge funds while managing a global team of analysts located in London, New York, Paris and Hong Kong.  He was a leading expert in FiCAD, a proprietary Computer Aided Design system developed in Objective-C (the core objective-oriented language that underpinned NextStep, a predecessor or Apple’s Mac operating system) by C*ATS Software, and helped develop the Value-at-Risk system while working at UBS in New York in the 1990s. At Société Générale he also developed recognised systems for measuring counterparty credit exposure for complex derivatives; The core system he designed- Exotix- was used by Societe Generale in North America for over 10 years.   Marc has in-depth knowledge of programming in C++, C# and Java and deep knowledge of the mathematics used in formal language theory.  He is a regular publisher on LinkedIn Pulse, which he will be exploiting to market to TOPOS® leads and clients.  Marc has developed all the core source code for TOPOS®’s products.  Finally, Marc has developed the Patent Pending methodology that will be used by TOPOS® to tag tweets for sentiment and mood. 

Galin Georgiev - Galin holds a Ph.D. in mathematics from Rutgers University and was a member of the Institute for Advanced Study, Princeton. Galin spent a number of years as a quant at J.P. Morgan and Bank of America.  As a Managing Director and a Board Member, he was a key player in developing a $10 Billion Alternative Investment fund - Pacific Alternative Asset Management Co -  where he managed a team of data scientists, quantitative analysts and traders.  During his tenure, he pioneered position transparency and built in-house a big-data/technology/portfolio management platform to process millions of position holdings. He subsequently created a successful high frequency trading firm, which was sold to a large hedge fund. He had also collaborated in the creation of a start-up “domain intelligence” platform for store sales forecasting and store relative value analysis, based on GIS, census analytics and consumer data. Most recently, Galin has concentrated on consulting in the field of data science, machine learning and deep learning with neural networks. Galin is fluent in SQL, R, Matlab, Python, Theano, C# and C++.
  1. IMAGE
  2. IMAGE