We are a trading firm based out of Beverly Hills, CA, and backed by the Chairman Emeritus of a major exchange on Wall Street. We are creating the next generation trading systems. Our technologies have impressed software legends such as Alan Kay and Doug Lea, and what we have been able to achieve with our small team have stunned giant trading firms like Merrill Lynch and Goldman Sachs.
Here are some of our current technical challenges, do any of them spark an interest for you? We are developing on Java 5 and Python. If you are interested in learning more, send me an email / resume highlighting experiences are relevant, let�s talk more and see where this will take us.
Grace Law email@example.com
1) Realtime OLAP/Materialized views. The OLAP demands of our system require near realtime updates to a number of datacubes summarizing our transactional data This cannot be accomplished by off-the-shelf systems (e.g. Oracle Materialized Views) due to the complex nature of some of the aggregations and other domain specific constraints, including auditability of presented views, propagation of summary data as transactional inputs to additional systems, and changes in the methodology used to aggregate and summarize the data (to reflect changes in the underlying financial models).
Tagline: Combine Oracle Materialized Views and Replication in a single flexible system.
2) Data integrity and anomaly detection Tracking the flow of cash and securities through our system is a complex process in and of itself, involving a large number of data sources as inputs and a larger number of rules regarding processing. However, the real challenge is ensuring the integrity of these transactions. Internal consistency is a must, but given the inevitability of "dirty" input (operator error or incorrect calculations from an external automated system) we must be able to quickly identify and remedy any anomalies.
Tagline: Building tools to ensure that a N-dimensional system of equations applied to large data set balances precisely, and when human error prevents it from balancing, identifying the source of the anomaly and recommending or applying corrective action.
Who would be good? The candidate must be comfortable digging through a large data set to determine what went wrong, and have the talent to automate the investigative process they just defined.
3) Systems engineering. Our system has a regulatory requirement for reliability. Aside from the challenges of making sure our system runs reliably, we also need to cope with an increasing number of users who may unintentionally damage our data. Most commonly, we deal with unreliable external data sources and we must be able to provide a wrapper that presents a reliable data source to our core systems. An example would be interfacing with an exchange that provides closing prices at a slightly different time each day, with occasional formatting differences, and with a potentially unreliable connection (e.g. a dial-up connection).
Tagline: Developing reliable and robust systems to interface with unreliable and fragile external systems.
4) Financial modeling. Our system implements and will implement a number of models for measuring and allocationg profitability and risk. The challenge is to apply these complex models to large datasets, presenting the data in multi-dimensional rollups as needed.
5) Workflow/Process management. Our system currently interfaces with a handful of highly skilled users who primarily care about detecting anomalies through the system. The challenge here is to determine the requirements from the users and implement them so that problems are identified as soon as possible and the proper tools are provided to research and remedy those problems. This challenge does not stand alone, but applies across all of the above areas.
Tagline: Better know how to talk to non-engineers.