The Role:
The right candidate will be able to research, spec and build the platform for semantic coding platforms in a funded startup. The candidate will become the an integral part of the business and the founders will look towards quickly building a team around this person.

Reporting into the CEO and CTO, this person will be a self starter, knowledgeable and confident in starting a project from scratch and making it their own, with the support of a highly adept team of marketing, operational and business experts.

Responsibilities include:
• Creation, programming and enhancement of AI based on semantic interpretation of large pharmaceutical and healthcare data sets
• Semantic interpretation task, combining techniques from across different fields such as statistical classification, information retrieval, stochastic parsing, machine learning and machine translation.
• learning, training models from large, unlabeled data and some of the work will focus on automatic, corrective learning based on user feedback.
• improvements to several components such as segmenting and document structuring, parts of language tagging, syntactic parsing, word sense disambiguation, semantic tagging and interpretation, normalizing extracted facts to standard code systems with or without underlying hierarchies and system confidence estimation.
• This will be conducted in the context of building a scalable, commercialisable healthcare extraction systems on top of available in-house resources and technologies - such as Ontologies, parsers, taggers etc
• There is also an opportunity to make the system robust to errors in system input, such those introduced by optical character recognition systems while converting image to text on uneven surfaces or in low light levels.

Qualifications:
Ph.D. in Computer Science, Computational Linguistics, Electrical Engineering or equivalent is preferred, but not mandatory.

Required Skills:
• Minimum of 5 years of experience conducting research in large statistical classification or statistical machine translation is required.
• Research experience in machine learning, stochastic parsing and semantic is a big benefit.
• Experience in semi-supervised and unsupervised learning methods, corrective learning from feedback during product use is also highly desirable.
• Programming experience in one or more of these languages is required: JAVA, C++, PERL, Python, Haskell, Epigram, Escher, JVM, F#.
• Bash/Unix scripting skills are required.

Preferred Skills:
• SQL or other database knowledge (preferably dynamic database knowledge) is highly desirable.
• Experience in medical text analytics and/or clinical information retrieval (such as pharmacy or medical systems/text) is a bonus.
• Experience working with UMLS components (MeSH, Specialist etc) is a bonus.
• Experience working with UIMA.