Software Development Engineer in Test II - Bing

Company: Microsoft
Location: Bellevue, WA
Web: www.microsoft.com/

Are you a highly technical, self motivated SDET who is looking for a bigger challenge? Are you tired of working on legacy releases? Do you want to put that theoretical knowledge you learned in college in good use? Do you want to write production code and experiment with new information retrieval techniques? Do you want to impact our stock price and go head to head with one of our biggest competitor? If you answer yes to those questions, read on...

In Core Ranking, our job is simply stated, "Match a query against billions of documents and produce the best ordering possible. Do this for thousands of queries every second. Do it really fast." Now as a tester in this area, your job is to ensure this is done well.

The impact you'll have on this team is extreme as you will be working on Relevance Test team in the Core Ranking space.

Here's a brief overview of Relevance Testing in Bing to whet your appetite:
1. This is not your traditional testing role - Typically, testing involves a lot of functional "does this work or not" type tasks. However, in the Relevance space, you have to answer questions like, "how relevant are we"? You have to measure and quantify our gain in Relevance. This is no easy task.
2. Data mining galore! - Since our job is about quantifying relevance, we also have to be excellent data miners. We have to know how to sample and aggregate data to ensure we are measuring ourselves and our competitors fairly.
3. Code, code, code - Every single person on this team is coding all the time. A search engine is so large and changing so frequently that you are required to automate nearly all of your testing work. We don't have time to do manual passes, so we have to ensure we have quick, thorough, highly technical software design engineers to build scalable automation suites that will withstand build after build, day in and day out.
4. Scale and size - The problems we work on are certainly not confined to one or two machines. We are working with hundreds of thousands of servers rolling out new pieces of the engine every single day. It is a problem solver's dream! How do we ensure quality in a virtually living system that changes so often? How can we cover all of the integration scenarios that will arise as a result? These are only some of the problems you'll face in this job.
5. Machine learning and Information Retrieval techniques at work - Typically, you study theory in college and then you work at a company doing very wrote, basic tasks. However, this is the place to apply all of that really fun knowledge you learned about AI, data mining, machine learning, and Information Retrieval. In the Relevance test team, you have to answer questions like, how does this new machine learning technique perform? Is it able to learn well given the data it is training on? What types of experiments can we do, be it cleaning data, manipulating learning parameters, or selecting different inputs, that will optimize for relevance and performance?
6. Diagnostic and debugging tool development - Some of the most important work we do on this team is targeted toward providing the team with a deeper view of 1) Data or 2) our production system. These tools drive and prioritize the development work we do, so you can''t get more of an impact than that.
7. Data Analysis tools - Part of our job is to find out the areas of improvements and finding patterns, analyzing win- loss of a technique by analyzing and mine through huge amount of data, applying statistical and machine learning techniques .

Job responsibilities:

• Improve core ranking algorithms by finding issues with ranking components across all of Search, training tools, and data
• Build tools to expose problems with training data and classifiers
• Improve testability in the relevance infrastructure components by participating in design review meetings and designing core components of your own
• Partner with Research to come up with new ways to test improvements in Relevance

• BS in CS or equivalent experience
• 5+ years designing and developing software
• Strong coding and debugging skills writing C/C++ or C#
• Strong analytical skills
• Solid communication skills verbal and written
• Ability to influence and drive quality across teams, across disciplines, and across organizations
• Ability to perform system-wide root cause analyses and drive improvements

• Machine learning and information retrieval techniques
• Data mining, statistical analysis experience
• Perl and SQL experience

_Contact_:
In order to be considered for this position, you must

apply online via our career site

If you have questions about this position, please contact:
eugenial at microsoft.com. Resumes received via email cannot be considered until you formally apply online.