Artificial Intelligence Client and Engine Development Required

I require the development of a basic Artificial Intelligence program, the program will need to work on Windows and should implement a form of AI that has been developed independantly. The project will require the programming of the AI engine, but full specifications on the functional requirements of the engine as well as a full description of the AI engines architectural specifications will be provided via skype (prefered) or email. The programmer need have no experiance in AI, though this would be preferred and need not work on the theoretical function of the AI engine, only its implementaion through code. The project has a budget of roughly $1000 and should take no less than one month, specifics regarding price/timescale are negotiable, so all offers are accepted. Thanks for reading and good luck bidding! Martin Williamson ## Deliverables The program will need to implement the following basic functionality; 1. Retrive raw information from a source (should be the moniter) and process this information into the AI engine. The program should be fast at this and it should be quite easy to have the engine process the screen and learn from it at a resonable rate (30-60 fps approx). The AI engines design will be provided, and the program will need to use this specification to store the captured screen information in a database or to save it in a file. Once this has been accomplished the program should be able to store a probabilistic model (full spec will be provided, but this is a rough outline) of the information that has been captured from the screen that will model the actual information that is displayed on the moniter. 2. The program will also need to implement a similar function as No.1, though it will need to be a keyboard/mouse WIN32API (or similar) interface that will allow simulation of keystrokes, mouse movements to the client computer - the same engine will be used, though in a reverse fasion, to probabilistically control the users mouse, while learning from information on the screen (as described in 1) which should allow basic probabilistic AI functionality (IE recognising the start menu and having seen the start menu clicked several times via screen capture, decide to click the start menu). The engine described should allow this functionality if implemented correctly - All the required neuroscience and technical details have already been researched and developed, so once the engine is operating under the engine as it will be described, this functionallity should show itself. 3. Full database/save file saving of the engine (both screen capture and interface simulation) so as to be able to load an engine that has previously been used to be used again from the same point and also to be able to save the engine so no processed information will be lost. 4. No bugs, fully developed and ready to run. The program should be in .exe format and include everything needed to be used - I have a proficiant technical background but I do not intend to have to do any work to bring this software to maturity, so it will need to be fully tested before submission. The engine design is in a complete state and Google Sketchup documention, mind maps and conceptual models will be provided. A small amount of work will be needed to turn these into a functioning engine, but this should only require programming skill and will not require any training in AI or neuroscience. A full book will be provided for background and eductional benifits to the winning bidder that explains in detail the principles of the engine and will greatly aid developent of the software, as the concepts underlying the AI engine will be fully covered (only one chapter of this book is useful and it may be required reading for the candidate programmer to fully implement the AI engine described in both the book and the technical specification documents provided by myself. As of yet, the functionality of output (keyboard and mouse) is conceptual - but the documentation covering the AI software to be programmed for the input should apply to the output as well - If both input and output work as intened, the software should be able to interact with the computer in an intelligent way. A small amount of conceptual discussion may be needed to implement the output functionality (I will explain the engine in full via email and the programmer should be able to utilize the specification to implement the output functionality in the same way as the input quite easily