Computer vision engineer to data scientist
A computer vision engineer already meets 26% of what the data scientist role asks for. The move turns on 46 required skills not yet in the profile.
83.4/ 100
1 How much of what a data scientist is expected to do is already part of a computer vision engineer’s work. Directional: the figure for the reverse move differs, because moving into a speciality is not the same problem as moving out of one. 2 Every skill the data scientist role is recorded as needing, required and optional together. What these numbers mean.
Why this move works
A data scientist role treats 17 of its required skills as things a computer vision engineer already does. You would walk in with these on day one.
- data engineering
- data science
- deliver visual presentation of data
- develop data processing applications
- establish data processes
- execute analytical mathematical calculations
What stands in the way is 46 required skills the profile does not yet cover. Table 3 groups them; Table 4 says which to take first.
Table 2 · What you already bring
Of the 24 skills that carry over, these are the ones fewest other occupations ask for. A data scientist role needs them, and most people applying for one will not have them already. This is the part of a computer vision engineer background worth leading with.
- already held develop data processing applications working with computers
- already held handle data samples information skills
- already held normalise data working with computers
- already held scientific computing information and communication technologies (icts)
- already held implement data quality processes working with computers
- already held perform data cleansing working with computers
All 24 carried skills, including the 17 the data scientist role treats as required.
Table 3 · What you may need to learn
The 73 missing skills fall into 11 areas of the ESCO skill hierarchy, numbered in the order worth working in. Areas carrying the most required skills come first, and inside each one the required skills sit above the optional ones. Nothing here is a verdict on you: it is a list of what this particular job asks for that your current one does not.
- still to learn design database scheme required
- still to learn manage intellectual property rights required
- still to learn communicate with a non-scientific audience required
- still to learn develop professional network with researchers and scientists required
- still to learn disseminate results to the scientific community required
- still to learn draft scientific or academic papers and technical documentation required
- still to learn evaluate research activities required
- still to learn increase the impact of science on policy and society required
- still to learn interact professionally in research and professional environments required
- still to learn mentor individuals required
- still to learn promote the transfer of knowledge required
- still to learn publish academic research required
- still to learn speak different languages required
- still to learn think abstractly required
- still to learn write scientific publications required
- optional, not held design database in the cloud optional
- optional, not held manage ICT data architecture optional
- optional, not held apply blended learning optional
- optional, not held teach in academic or vocational contexts optional
- still to learn apply for research funding required
- still to learn apply research ethics and scientific integrity principles in research activities required
- still to learn collect ICT data required
- still to learn conduct research across disciplines required
- still to learn demonstrate disciplinary expertise required
- still to learn integrate gender dimension in research required
- still to learn manage findable accessible interoperable and reusable data required
- still to learn manage research data required
- still to learn perform scientific research required
- still to learn promote open innovation in research required
- still to learn promote the participation of citizens in scientific and research activities required
- still to learn synthesise information required
- optional, not held manage data optional
- optional, not held make data-driven decisions optional
- still to learn data mining required
- still to learn data models required
- still to learn information extraction required
- still to learn online analytical processing required
- still to learn data visualisation software required
- optional, not held Hadoop optional
- optional, not held LDAP optional
- optional, not held LINQ optional
- optional, not held MDX optional
- optional, not held N1QL optional
- optional, not held SPARQL optional
- optional, not held XQuery optional
- optional, not held unstructured data optional
- still to learn build recommender systems required
- still to learn operate open source software required
- still to learn manage open publications required
- still to learn use data processing techniques required
- still to learn use databases required
- optional, not held integrate ICT data optional
- optional, not held manage ICT data classification optional
- optional, not held use spreadsheets software optional
- still to learn empirical analysis required
- still to learn scientific literature required
- optional, not held multidisciplinary research optional
- optional, not held research design optional
- still to learn information categorisation required
- still to learn quantitative analysis required
- optional, not held digital curation optional
- optional, not held social network analysis optional
5 further areas in the appendix
Table 4 · Where to start
The 3 entries a data scientist role is least likely to hire without. The ordering is computed from the skill data, not from what pays.
Each entry opens a course search for that skill. Career Overlap earns nothing from these links.
Appendix · The rest of the record
All 24 skills that carry over
- already held data engineering knowledge
- already held data science knowledge
- already held deliver visual presentation of data skill
- already held develop data processing applications skill
- already held establish data processes skill
- already held execute analytical mathematical calculations skill
- already held handle data samples skill
- already held implement data quality processes skill
- already held interpret current data skill
- already held manage data collection systems skill
- already held mathematical modelling knowledge
- already held normalise data skill
- already held perform data cleansing skill
- already held query languages knowledge
- already held report analysis results skill
- already held resource description framework query language knowledge
- already held statistics knowledge
- already held computer simulation knowledge
- already held create data models skill
- already held define data quality criteria skill
- already held image recognition knowledge
- already held perform data mining skill
- already held scientific computing knowledge
- already held state estimation knowledge
The 5 learning areas not shown above
- still to learn visual presentation techniques required
- still to learn data ethics required
- still to learn manage personal professional development required
- still to learn perform project management required
- still to learn statistical modeling techniques required
- optional, not held business analytics optional
- optional, not held computational biology optional
- optional, not held data quality assessment optional
- optional, not held business intelligence optional
- optional, not held marketing analytics optional
- optional, not held healthcare analytics optional
27 supplementary skills, helpful but not required
- optional, not held digital curation optional
- optional, not held Hadoop optional
- optional, not held LDAP optional
- optional, not held LINQ optional
- optional, not held MDX optional
- optional, not held N1QL optional
- optional, not held SPARQL optional
- optional, not held XQuery optional
- optional, not held business analytics optional
- optional, not held computational biology optional
- optional, not held data quality assessment optional
- optional, not held design database in the cloud optional
- optional, not held healthcare analytics optional
- optional, not held integrate ICT data optional
- optional, not held manage ICT data architecture optional
- optional, not held manage ICT data classification optional
- optional, not held manage data optional
- optional, not held unstructured data optional
- optional, not held apply blended learning optional
- optional, not held business intelligence optional
- optional, not held make data-driven decisions optional
- optional, not held marketing analytics optional
- optional, not held multidisciplinary research optional
- optional, not held research design optional
- optional, not held social network analysis optional
- optional, not held teach in academic or vocational contexts optional
- optional, not held use spreadsheets software optional
28 held skills the data scientist role does not ask for
- not needed by the target role Python (computer programming) knowledge
- not needed by the target role apply statistical analysis techniques skill
- not needed by the target role cognitive computing knowledge
- not needed by the target role computer graphics knowledge
- not needed by the target role computer programming knowledge
- not needed by the target role conduct literature research skill
- not needed by the target role conduct qualitative research skill
- not needed by the target role conduct quantitative research skill
- not needed by the target role conduct scholarly research skill
- not needed by the target role debug software skill
- not needed by the target role deep learning knowledge
- not needed by the target role define technical requirements skill
- not needed by the target role design user interface skill
- not needed by the target role develop computer vision system skill
- not needed by the target role develop software prototype skill
- not needed by the target role digital image processing knowledge
- not needed by the target role digital systems knowledge
- not needed by the target role digital twin technology knowledge
- not needed by the target role image formation knowledge
- not needed by the target role integrated development environment software knowledge
- not needed by the target role machine learning knowledge
- not needed by the target role perform dimensionality reduction skill
- not needed by the target role principles of artificial intelligence knowledge
- not needed by the target role quantum computing knowledge
- not needed by the target role signal processing knowledge
- not needed by the target role use markup languages skill
- not needed by the target role use software libraries skill
- not needed by the target role utilise computer-aided software engineering tools skill
30 gaps that are knowledge rather than practice
Knowledge gaps usually close through study. Practical skill gaps usually need something you can point at.
- still to learn data mining required
- still to learn data models required
- still to learn information categorisation required
- still to learn information extraction required
- still to learn online analytical processing required
- still to learn statistical modeling techniques required
- still to learn visual presentation techniques required
- still to learn data ethics required
- still to learn data visualisation software required
- still to learn empirical analysis required
- still to learn quantitative analysis required
- still to learn scientific literature required
- optional, not held digital curation optional
- optional, not held Hadoop optional
- optional, not held LDAP optional
- optional, not held LINQ optional
- optional, not held MDX optional
- optional, not held N1QL optional
- optional, not held SPARQL optional
- optional, not held XQuery optional
- optional, not held business analytics optional
- optional, not held computational biology optional
- optional, not held data quality assessment optional
- optional, not held healthcare analytics optional
- optional, not held unstructured data optional
- optional, not held business intelligence optional
- optional, not held marketing analytics optional
- optional, not held multidisciplinary research optional
- optional, not held research design optional
- optional, not held social network analysis optional
Common questions about this move
Can a computer vision engineer become a data scientist?
Yes, and it is a career change move. A computer vision engineer already covers 26% of what a data scientist is expected to do: 24 skills carry over, 17 of which the data scientist role treats as required.
What skills does a computer vision engineer need to become a data scientist?
46 required skills are missing, plus 27 optional ones. They fall into 11 areas of the ESCO skill hierarchy, led by communication, collaboration and creativity. The single most important is build recommender systems.
How hard is the move from computer vision engineer to data scientist?
Its learning distance is 83.4 out of 100, which is the career change band. That accounts for both how much is missing and how specialised those skills are.
How this record was compiled
Both occupations are taken from ESCO, which lists the skills and knowledge each occupation is expected to have and marks every one required or optional. Nothing here is a prediction about hiring, and nothing here knows that a particular employer wants a particular certificate. Treat Table 3 as a starting point for your own research rather than a syllabus. The full method states what these figures can and cannot tell you.