We care deeply about:
Engineering excellence such as good design abstractions, API stability, best practices, and pragmatic testing
The incremental approach to providing customer value
Removing technical debt and finding ways to improve an existing system while managing risk
What You'll Do
Utilize your engineering and machine learning expertise to develop models and solutions that improve Twitter’s geolocation services and data sets.
Partner with different teams throughout Twitter. You’ll collaborate with engineers, ML modellers, and product teams to build high performing systems that have impact.
Help drive the roadmap and determine the best way to leverage ML.
Apply your expertise in distributed systems to simplify the complexity of wrangling large datasets.
5+ years industry experience with a Bachelors, MS or PhD in computer science, data mining, machine learning, information retrieval, recommendation systems, math, engineering, operations research, or other quantitative discipline; or equivalent work experience
Experience in building production ML models and understanding of inference challenges at scale
Good theoretical grounding in core machine learning concepts and techniques
Experience with a number of ML techniques and frameworks, e.g., data discretization, normalization, sampling, linear regression, decision trees, SVMs, deep neural networks, etc.
Familiarity with one or more deep learning software frameworks such as Tensorflow, PyTorch
Backend development experience with the ability to work on data pipelines, distributed systems, performance analysis, and/or large-scale data processing
Experience with software engineering practices (e.g. unit testing, code reviews, design documentation)
Who We Are
The Geo team maintains ownership of our notion of user’s locations, points of interest, and places at Twitter. We work with large datasets and high traffic systems while protecting our user’s privacy.
We aim to help Twitter deliver great content to each user by understanding the specific places they are interested in and identifying the right content associated with each place.
Our engineers build services and tools to automate and simplify our current Geo operations, including ingestion pipelines, location calculations, and other user-facing improvements to our Geo systems.