A masterclass in Scientific CV writing

Introduction: Another day, another application form

Writing applications for jobs, grants and all manner of other reviews is a continual process within the scientific World. Forms tend to ask for specific, nuanced information leading to more of our precious time being spent digging up decades-worth of buried events just to evidence ‘A time I have communicated with a diverse audience’ than actually writing. Then, we have the doubt to contend with: What if I missed something? Surely I have a better example! I remember doing that – but when was it?

Given A) how short academic contracts can be and B) how many distinct workplaces our generation tends to work in over the course of a career, writing CVs can consume a considerable chunk of our adult lives. The application process is not going anywhere in the near future. We need to ask ourselves how we can make it as painless and efficient as possible.

Well, there are a few ‘hacks’. Apply for a few jobs and you will start to notice themes in the application process and in the ‘winning’ CVs. Let’s go over these themes and learn to not only ‘hack’ our time but more importantly, our success rate. Doing so, we can earn back so much more time to do the things we love – science!

1) Maintain a private, point-by-point autobiography of your professional life

We will call this your private ‘CV’ though, an exhaustive personal record of every achievement, experience & skill would be more accurate. When you need to send a CV for a specific position or grant, all you will have to do is copy, paste and mildly adapt (I’ll call this ‘patch’ to keep with the handiwork theme) from this private CV.

Those open days you worked as an undergraduate? Add them!

The time your research was featured on the faculty poster board? It’s in!

You helped a colleague with using a software package? You get the point. Each of these experiences evidence your engagement, communication, competence and proactive nature – things you will undoubtedly need specific examples of.

Key CV point A: As a general rule, if something took you more than five minutes to learn or to do, it was a development experience. Record it!

This is going to be a long document. It will involve an initial investment of 1+ work days to write but, the amount of time you will save throughout your career will be many times more than the initial investment. Given how valuable this document will be, make sure it is not lost! That means:

Key CV point B: BACK UP, BACK UP, BACK UP!

The 3, 2, 1 rule applies here – 3 copies of the document on at least 2 drives, 1 of which is at another location. I ‘hack’ this process too – I have a private ‘CV’ repository on GitHub (free backup to ‘the cloud’ – copy 1 at location 1) which I clone to both my laptop (copy 2 at location 2) and workstation (copy 3 at location 2). To mitigate for errors in GitHub, I also upload a copy of this CV after every editing session to my Google Drive (OneDrive, DropBox etc. will do the trick too, as long as it’s a different cloud solution to the GitHub repo).

Key CV point C: Name your private CV file unambiguously!

Whatever you do, please mitigate the risk of accidentally sending your private CV to the application committee. A 57 page autobiography is a direct route to HR’s naughty list. Name the private CV file in such a way as to prevent future you from ever considering sharing with another living soul.

DO_NOT_SEND_ME_Private_CV_FOR_MY_EYES_ONLY_DO_NOT_SEND.docx’

should do the trick.

Key CV point D: Update it fairly frequently!

How often you add content to the private CV depends on the events in question. If you publish a paper, give a talk or learn a new skill, those should be added to the private CV as quickly as possible for the simple reason that they are fresh in your memory. Specifics such as attendance numbers, co-author names & URLs are highly valuable for the CV but become fairly hard to find after a few weeks.

For the larger scale content of the CV -objective, experience summary, professional experience contact details etc. – these will likely stay relevant for some time (unless you have a sudden change of heart…). Adding new experiences and making minor adjustments to the above every 6 months should be sufficient. To make sure this gets done,I created a recurring calendar event. I now have dedicated, scheduled time to focus just on my CV.

Given the pace we work at, calendar reminders are our best friends in ensuring we stay on course.

2) Supercharge your CV with specifics!

The purpose of a CV is to convey who you are in as little space as possible. Detail is going to make it crystal clear whether you are right for a role and more importantly, whether the role is right for you. When writing your private CV, make sure to record all those valuable, character-revealing details so easily forgotten later.

Let’s start with what we’d like to avoid.

The commonly-used phrase ‘I train deep learning models’ is ambiguous. While it certainly reveals which large area of ‘science’ we are in, it doesn’t narrow things down as much as needed for a CV. This is especially true now – an age where an ‘artificial intelligence focus’ is actively promoted in, by my assessment, most projects. Never before have there been so many avenues in the deep learning field. This statement doesn’t reflect who we are, as deep learning practitioners, and where we would be suited in such a vast space.

Now let’s add that special detail sauce:

‘I trained a U-Net model for the segmentation of epithelial cells on digital whole slide images, improving sensitivity from 0.73 to 0.85 using Python 3, Sci-kit Image, PyTorch, and Albumentations.”

Why is this better than the earlier example?

Your chosen branch of DL

Within deep learning, there are a plethora of broad topics: natural language analysis, sequence analysis, image analysis and so on. Within each are countless further subdivisions, carrying niche-specific nuance at each branch. Specifying the nature of the problem tackled; model architectures used; input data format and your role within it pinpoints your position within the grand tree of deep learning.

The benefits here are two-fold:

1) you are providing credibility – if someone can specify what they do, it evidences that they know what they are doing

2) you are helping recruiters to assess the distance between your skills and the desired work. Making their lives more comfortable pays dividends, of course, but keep in mind too that recruiters need to hire for multiple positions – if you have been clear about your specialisation, you will be at the top of their list when a similar position comes up.

You have clarified your programming language; deep learning framework of choice and specific libraries

This is key for assessing ‘fit’ in future workplaces. Whether a lab is Python-ic, Rust-acean, cRan-ical or other speaks of the culture and approach.

Likewise, whether a team builds in TensorFlow or in PyTorch is informative of whether deployment or development are their aims.

Finally, the libraries used evidence your mindset and focuses. Having experience with Albumentations evidences an appreciation of data augmentation and model generalisation – this may be a central theme of a project so, being specific to this level boosts your position in the candidate list. Ensure a smoother transition into the right workplace by unambiguously communicating your programmer and thinking style!

You have explicitly delineated your particular added value to the project

This need not be something quantitative but, giving a sense of how the project began before you and was bettered by your involvement, as evidenced by how it was left, is critical in demonstrating the value you bring to projects you are engaged in. For example, you can say “reducing turn-around time of approvals by installing an automated system” is better than “installing an automated system” because it clearly demonstrates the value your efforts converted into. That said, quantitative metrics are the gold standard – “reducing turn-around time of approvals by 14%, on average, through installation of an automated system”… perfection!

Numbers speak volumes

Numerical values are superb for evidencing the extent of your impact. Anyone worth their salt in the deep learning community knows that a sensitivity improvement of 0.12 is fantastic. This number grabs a recruiter’s interest more effectively than words alone.

Honestly though, most things we do may not have such a large impact. There’s no need to feel downcast about this. Including figures to evidence impact is great even if your achievements seem modest. It subtly reveals a precise, methodical & quantitative attitude – all essential qualities for sustainably conducting scientific work. So, avoid generalisations and sweeping statements on your impact – let the numbers do the talking!

3) Start with the finish – structure with purpose!

As a teacher in a previous life, I would maximise my efficiency by planning lessons backwards from my desired outcome. The same ‘hack’ can be applied here.

You can minimise the time needed to construct CVs specific for a role by organising your private CV into commonly required subsections! Now, this section is going to be a bit long. Please don’t fret. Follow this format and the ease of transfer to a job-specific application will speak for itself.

Part 1: Your background

Note: for anything involving time, always start with the most recent experiences and work backwardsin time.

A) Contact:

Let’s give recruiters a list of ways to 1) get in touch and 2) do some further research on our professional histories. That means our name, professional phone number, professional email address, GitHub page, Institution page, LinkedIn page, ORCID ID etc.

B) Objective:

Here, we will write two sentences outlining who we are, professionally, and where we are headed. Not only an effective way of ensuring we are ‘pigeon-holed’ for roles suited to our ambitions, it is also a great way to remind us of what we are all about. It’s easy to forget our motivations and apply en-masse for roles that do not align with your values or dreams (I am guilty of this). Let’s save grief and time – self-reflect, write a clear personal outline and be true to it!

Mine would be:

A computer vision practitioner with 2 years of experience in automation of histopathology image analysis in lymphoma, focusing on features predictive of patient outcome. My career direction is in development of Python pipelines to inform treatment choice and in dissemination of such software to clinicians via the Flutter app development framework.

C) Experience summary

Here, we will make a short, concise bullet point list of four to six key, niche skills we have acquired. This is going to complement our objective above by providing a lot more context, in a space-efficient way. To make clear our level of experience in each, include the years of practice in bold. An example summary (not mine!):

  • Natural Language Processing (3 years)
  • Sentiment Analysis (2 years)
  • Sales prediction (German market; 1 year)
  • Python Programming (4 years)
  • TensorFlow Deep Learning Framework (2 years)

It’s clear that the above belongs to someone from a deep learning for business background, possibly using German language review or social media data to gauge buyer experience and more recently, to predict buying habits. This clearly puts into perspective suitability for a role and recent trajectory.

D) Education:

A list of formal, higher education qualifications such as Bachelor’s degrees, Master’s degrees, PhDs and MDs. In our digital era, what constitutes ‘formal education’ is more of a grey area than ever. See the ‘Online courses’ section below for clarification. Include the degree course, the qualification, module grades, overall grade and dates of study.

E) Professional experience:

Here, we will list every role we have held, responsibilities within those posts and crucially, which skills we developed. Let’s not forget to include the dates over which we held these positions.

F) Online courses:

The distinction between online learning and formal education is less clear in this decade. Many massive open online course (MOOC) providers now offer accredited graduate programmes – it has never been easier to be misinformed about the professional extent of online qualifications.

We will reserve this section for online courses that did not award university (college, for our pals across the pond) credits. Some platforms make the accreditation of courses clear while others offer ‘professional certificates’, the validity of which is anyone’s guess.

As a rule, err on the side of caution. If in doubt about the accreditation of an online course, include it in ‘Online courses’ rather than ‘Education’. It is better to appear modest yet motivated – pursuing study in your free time! – over having HR discover that your qualification from MIT is not as endorsed by MIT as the MOOC page suggested.

Besides, your professional experience and output – how you applied what you learned! – will speak much more loudly about your skills than the formal designation of an online course.

Part 2: Your output

G) Citation statistics

Let’s summarise our entire literature output here. We can use Google Scholar to retrieve much of this data – citation indices (with date you retrieved this figure on and which provider i.e. Google Scholar), number of citations, h-index and i0-index.

H) Journal papers:

The coveted product of any academic research project. Here, we will cite (in one of the popular formats – Harvard referencing is a good go-to) complete research articles that have undergone peer review and been accepted into a journal. Let’s make our level of contribution clear by formatting our name in bold andour work easy to find by including a PubMed ID.

I) Preprints:

Using the same format as above, reference your works that have been submitted for publication, have been published to a preprint server (the ArXiv family, AAS Open Research, Thesis Commons etc.) but have not been published by a peer-reviewed journal, yet. It would be best to include the submission date and publication status here, also.

J) Conference contributions:

In this section, we will reference work presented at events attended by members of our field. We can follow the same format as set out in the ‘Journal papers’ section above with the addition of abstract / paper / presentation in brackets (parentheses) to ensure accurate reporting of the format.

K) Chapters:

Cite, as above, any chapters written for academic books. If you have had the fortitude to write an entire book (!), cite that in a separate section above this one.

L) Grants:

The bittersweet fruit of the academic tree. List grants you have been involved in the writing of. Make sure to outline whether they have been A) Granted and completed B) Granted and ongoing C) Pending funding D) Submitted but not accepted, yet.

M) Patents:

Here, we will list patents we have been involved in. Required details would be patent name, patent number, application number and, if provided, the issue number. As for grants, it’s advisable to include the progress of the patent filing process – A) Granted and expired, B) Granted and ongoing, C) Pending approval.

N) Invited lectures:

These are a tad different to a talk or seminar as the purpose is to teach rather than to present and discuss. They are most often given as part of a University (College) module – the University has recruited you as your capabilities are deemed exceptional in that field. You may have given an invited lecture entitled ‘Epigenetic control in humans – nature, dependent on nurture’ to a group of second year undergraduates (sophomores). An invited talk, on the other hand, may have been much more specific to your topic (i.e. ‘Epigenetic control of TP53 in haematological malignancies’) and would likely be to a group almost as familiar with the area as yourself (i.e. an equivalent department in another University). The reason we have separate categories for lectures and talks (section O, below) is that there are different skills at play when communicating ideas to people within and outside of our area.

In this section, list talks you have given that qualify as invited lectures. make sure to include the department, establishment / organisation, topic and date. Why not demonstrate your impact by recording the number of and speciality of the attendees?

O) Invited talks:

Use the guidelines above (section N) to figure out whether your talk was a talk or a lecture. If it qualifies as a talk, record it here. Note down the same details as for an invited lecture.

P) Demonstrations:

Here, we will list times that we have demonstrated usage of some specialist tool. The point of a demonstration is to sell – metaphorically and / or literally – your product or your technique. Think back – have you presented a technique, such as use of a particular mass spectrometer, or maybe a software package you created and published with the intention of persuading others to use it? If so, list it here. Include the same information as for ‘Invited lectures’ (section N, above).

Q) Training sessions:

These differ to a demonstration in that the focus is to teach rather than persuade – you wanted people to understand how to use a tool. If you have conducted a session to teach people to use something, whether software, hardware or a humble piece of lab equipment, list it here. Include the same information as for ‘Invited lectures’ (section N, above).

R) Industry collaborations:

Keeping in mind non-disclosure agreements (NDAs) and sensitivity, provide details of the nature of partnerships with industry or commercial partners. If you are not privy to discuss the exact topic then detail the general area. For example, my NDA with GSK limits discussion of my exact work but I am at liberty to say that I worked in the Respiratory department, focusing on immune responses. It’s not much but it hints at my skills while paying due respect to the intellectual property owner. It’s best to agree with your industry contacts beforehand what you are and aren’t able to discuss.

S) Awards

Examples would be podium positions for conference posters; recognition for teaching excellence or success in a piece of scientific writing. Include topic, date and context for any formal recognition received for your work. If a URL is available, that’s superb – include it, too!

T) Media Recognition:

Media recognition could be a mention of your work on a local radio show; an interview with reporters or even a television appearance. Basically, any time dreams of Hollywood have seemed a fleeting possibility. Report it here (pun intended), including the same information as for awards (section S).

U) Service:

Beyond the scope of your roles, have you taken on any further responsibilities? Here, we could include any committees we have sat on; events organised or initiatives engaged with. You can outline these roles in quite a bit of detail within the private CV, including almost every contribution you have made. When it comes to the job-specific CV, you can cut-and-paste whichever experiences are relevant.

V) Memberships:

List any memberships to approved professional organisations and learned societies. In the UK, there is a government website for finding out if an organisation is one of these (can be found by searching for: ‘Approved professional organisations and learned societies (list 3)’). To determine whether the organisation you are an associate, member or fellow of is an approved one, please refer to such government sources in the country of origin of the organisation. If the organisation is not approved, it would be best to pivot and instead record your contributions to the organisation under the professional experience (section E), service (section U) or industry collaborations (section R) subsections. Which one would depend on the nature of your involvement.

Include the date range for which the memberships were held and the level of membership (associate, full member, fellow, honorary?). If you had any responsibilities within the organisation (i.e. presented invited talks or organised conferences), include those in the relevant sections above.

W) Open source contributions:

Did you make a successful push request on a GitHub project? Are you a StackOverflow Guru? List out your contributions and ideally, impact metrics, too – number of stars, forks, upvotes. As with citation metrics, make sure to include the date you recorded those metrics.

Part 3: Your contributions to the work of others:

X) Editing / editorials / reviewer:

Here, we will outline where you have provided feedback for the work of others in an official capacity, demonstrating your understanding of the field. This could include editorials or even instances where you have been part of the peer review process (i.e. an editor). For reviews you’ve completed, it is typically better to say “reviewer for journal XX” as opposed to listing the titles of the articles reviewed, to retain anonymity. Some journals are moving towards complete transparency in reviewers. Whether you list the article or the journal here depends on the individual focus of the journal in question.

Y) Supervision:

Can you include the work of those you have supported? Absolutely!

In this section, you can list out the people you have supervised and / or mentored.

Make sure to reference student theses that have been written under your supervision. Also, include any awards won by those students. For your private CV, make sure to include the student first name, year of supervision, topic and any other details needed to jog your memory if asked to ‘describe a time you have supported the work of others’. For the purposes of privacy, highlight any identifiable information in yellow so it’s easy to spot and delete when constructing the job-specific CV, unless you have the explicit consent of that person to mention them.

Part 4: Skills, skills, skills

Z) Technical skills:

List out all tools you have learned to use. This is primarily for us to not forget all the valuable experience we have gained (easily done at our pace of work). As stated, this can include anything that has taken longer than 5 minutes to learn. You may want to include details of context or specific task to jog your memory. An example, with the styles of notes you could add:

  • Docker – Containerisation of SynthGene and BubbleSorter for release
  • GitHub – daily project version management
  • Travis CI – testing Michael Walker’s CNN-Explainer tool (November 2021)
  • Micro-pipetting – undergraduate Biochemistry labs
  • SnakeMake – gene expression filtering
  • Anaconda / Miniconda CLI – daily package management; sharing and reproducing environments
  • ProxMox – recycling my old laptop as a NAS and VPN
  • Pritunl VPN – setting up my home VPN
  • Open Media Vault – setting up my home NAS
  • HomeBrew – package manager for some of the tools used in MacOS
  • Bash – daily automation of command line tasks
AA) Miscellaneous skills:

A list of every other skill you have, in case you need to evidence qualities demonstrated in these. Why not include details of your experience, too? Examples:

  • Chess (since 2017)
  • Driving (full licence obtained in 2013)
  • Hiking (Mount Snowdon scaled twice; Pennine Way planned for 2023)
  • Motor vehicle repair (started in 2020; progressed from minor, detail maintenance to most service tasks inc. spark plug, brake fluid, oil filter, cabin filter, brake pad and coolant changes, 2022)
AB) Languages:

For those of you that read this far, thank you 🙂

Let’s list out languages we know / have learned and an honest view of our proficiency. As discussed later, we need to be honest here. More so to save ourselves the embarrassment of being completely out of our depth. One of my colleagues from a past life, teacher training, oversold his proficiency in Welsh. For context, Welsh is an ancient language that predates the Latinisation of British society – it is very different to modern English. The impressed programme administrators assigned him to teach in a Welsh-speaking school for the next week. He had to hastily learn a language he had largely forgotten over 11 years of speaking English. Then, he had to teach fairly specialised content to native speakers. I salute his courage and resilience. He revised the language skills section of his CV fairly quickly afterwards.

Being outside of your comfort zone is an essential part of the learning process. If you aren’t as skilled as you would like to be in a language, it’s not something to be ashamed of or hide. If applying for a position that has language requirements, make it clear to the recruiter that you recognise this as a development point and evidence your enthusiasm by referencing efforts to improve. For example, if using an online platform like Duolingo, note down your usage frequency and streak length (make sure you can evidence this through the App), proficiency level and some recent topics covered. Like with citation statistics, record the date that you obtained these figures.

4) The most important policy is truth

Finally, the core assumption of a CV is that it is entirely accurate. Inflation, exaggeration or augmentation of any detail will violate that trust and bring into doubt everything involved. This includes changing the order of authors in papers or even rounding your h-index. Most employees, understandably, have immediate dismissal policies for such transgressions.

This doesn’t mean you should come across as any less capable than you are, just, be honest about times things didn’t work out as well as hoped or you didn’t contribute as much as would be implied on the surface. Prepare instead for how you moved forward from that event.

For example, let’s say you were part of a committee to organise an internal conference.

Other responsibilities got in the way and you had to leave this role. You may be tempted to state on your CV that you were a part of the committee and leave it there. That would be somewhat dishonest as the implication is that you served the full term from inception to delivery.

Now, by A) adding the months you served on that committee and B) pointing out your precise contributions, you can more accurately report your input. Myself, I would actually hope that this is questioned at interview so I could discuss how critical evaluation of my commitments led to the conclusion that my time and attention were required more elsewhere. Suddenly, we’re demonstrating critical thinking, analytical decision making and effective time management.

I would go so far as to say that our response to failure – objective, candid evaluation followed by reflective, reasoned revision of our plans – make us far better candidates for scientific work.

In summary, be clear of the facts in your CV and focus on the development you underwent as a result of disappointing outcomes. Where skills, experience or grades may be a ‘tad light’ for a particular role, outline to the recruiter how you are actively bettering yourself. By being open with recruiters, you will subtly evidence far more valuable character traits.

Outro:

So, there we have it, folks. A roadmap for time-efficient creation of a winning CV. Your behemoth of a private CV will contain all the information a job application, performance review or other, not yet dreamt of application process may require. You will no longer need to spend days in furious remembrance of that thing you once did which may or may not demonstrate said skill. Simply:

  1. find a relevant section of your private CV
  2. copy and paste it to a new document
  3. delete content not relevant for the role in question
  4. make minor edits to help the ‘flow’

Point evidenced. Rinse and repeat for the remainder of the ‘essential’ and ‘preferred’ skills.

Happy writing and best wishes for your applications!

Note on author and contributions:

Volodymyr Chapman is a 3rd year PhD student at The University of Leeds, UK, using deep learning on digital pathology data to predict outcome in non-Hodgkin lymphoma.

This article was written as a combined effort by Volodymyr Chapman and Andrew Janowczyk. VC conducted writing of the post. AJ conceived the idea for the post; consulted on content and provided feedback during writing.

Leave a Reply

Your email address will not be published. Required fields are marked *