Jobs at Eddy Systems

This is how we program today: 1. Human dreams up vague, intuitive code. 2. Human laboriously translates vague code into low level programming language. 3. Computer sees only (2). It provides no help with (1) or the move from (1) to (2). We want to fix this. Programming should be a conversation, with the computer joining the conversation as soon as possible. Instead of humans pretending to be computers (simulating code in their own minds, memorizing all the details of an API, etc.), we want to free ourselves to write incomplete and ambiguous code and have the computer talk with us to pick up the pieces. Eddy will be an interactive programming system that fills in details automatically, asks questions when it can't, and lets you choose how much detail to see as you read and modify code. Think iPhone autocorrect, but using types and inference to understand what you mean, not just fix spelling errors. To make this possible, we will be co-designing the user interface and reasoning engine together, outputting plain Java for full compatibility with existing ecosystems. The ability to ask the user questions is amazingly powerful but weirdly foreign to the world of languages and compilers, which must work automatically and are never allowed to make mistakes. For more information about what we want to achieve, watch our concept demo: * eddy.systems About us: Martin Wicke received a Ph.D. in Computer Science from ETH Zurich, then worked as a Postdoc at Berkeley and Visiting Professor at Stanford working on physical simulation, computational geometry, and sensor networks. Most recently, he led the Otherlab Compute team, building software for interactive CAD / CAM including a collaborative online voxel editor. Geoffrey Irving received a B.Sc. in Mathematics and Computer Science from Caltech with an emphasis on languages and compilers, and a Ph.D. in Computer Science from Stanford on physical simulation. He has worked at Pixar, D. E. Shaw Research, Weta Digital, and Otherlab, and received film credits on