The Student Room Group

This discussion is no longer active so you won't be able to reply.Check out other Related discussions

Working as a Data Analyst/Data Analytics

Hello TSR

I would like know about anyone's experience working as Data Analyst or in the area of Data Analytics. I have 1st class degree in chemistry from RG university with a good grounding with SQL server/scripting.

Look forward to your responses.

Cheers.
Are you thinking big business or smaller startups? The roles will be substantially different. What math background do you have? Do you have a preference for online tech vs finance vs engineering vs science orientated companies?

let me know a bit more and I'll try give you more insight.
Reply 2
Original post by andrewishiring
Are you thinking big business or smaller startups? The roles will be substantially different. What math background do you have? Do you have a preference for online tech vs finance vs engineering vs science orientated companies?

let me know a bit more and I'll try give you more insight.


Hi,
I am also trying to break into the world of data science but finding it hard to get my foot in the door as I do not have direct technical experience with regards to the machine learning algorithms,finding trends in data sets and data visualisation techniques etc but this is what I am interested in and hence trying to do self learning. My background is in healthcare - I have been working as a scientist in NHS and some other private companies for a number of years so have done data processing etc but there is usually specialised software or in-house programs to do it (althouhg I do know how to code in a few languages at a basic level). I have been trying to do some self -learning but finding it difficult to know what to expect in these roles.

Would you suggest aiming for startups or SMEs if one is trying to gain experience ? Ive heard it can be quite stressful with the workload. Im looking for somewhere where I can slowly develop myself with less customer/consultancy/client focus and more on the coding/analytical side (salary is not too much of a concern for me at the moment). Any insights would be helpful. I have seen a few healthcare startups recruitng in this area but not sure what to expect. Do you know much about the public sector ?

Thanks in advance
Original post by rk1103
Would you suggest aiming for startups or SMEs if one is trying to gain experience ? Ive heard it can be quite stressful with the workload. Im looking for somewhere where I can slowly develop myself with less customer/consultancy/client focus and more on the coding/analytical side (salary is not too much of a concern for me at the moment). Any insights would be helpful. I have seen a few healthcare startups recruitng in this area but not sure what to expect. Do you know much about the public sector ?


I have no experience with the public sector, so all these comments relate to the private sector - specifically startups with between 1-35 people.

You say you've been working for a number of years. What is your concern regarding work load in smaller companies? I would say that if you're looking to learn while changing roles, then you're probably going to have to deal with more than a standard 9-6 job. In most smaller companies there is more work than time simply because the company is trying to figure out how to grow and survive. The people working in them typically love what they do and hence work long hours because it's the only way a startup becomes successful. In larger companies the long hours come from different pressures and the rewards for success are different. This is why someone people love them and some prefer small companies. The work is also different because startups are by definition trying to find market fit rather than executing on a going concern.

The benefit of all this work means that most smaller companies are open to hiring people who may not cost much but who can help them. They will usually tolerate learning on the job if you're not costing them too much - money or time wise. But you'll get exposure to SO MUCH STUFF that if you're s sponge, you'll learn loads and probably open yourself up to other options.

The data science industry is in a massive state of flux as it defines itself. Almost all business can benefit but expert skills are hard to come by - especially while everyone seems to be selling snake oil in terms of skills. This is a double edged sword as some people expect the world while others are understanding of the current state of play and that it's not a form of magic that will solve their problems over night.

Most data science roles are best filled by people who have (or can acquire quickly) a very good understanding of how the business works and what makes it successful. A lot of skills in data science are transferable as long as you understand how to apply them in a different business.

I think you should be able to find some smaller companies who may not be actively looking for someone like yourself but who could benefit from your expertise and desire to learn. If you can afford to, I'd suggest trying to offer your skills at a discount while you skill up. Find an organisation that has a similar set of problems to the ones you've been solving. This doesn't mean the same industry. For instance, maybe you were looking for correlations in dataset of patients at the NHS. This should apply just the same to finding correlations in e-commerce audiences.

I assume you're young. Take advantage of this and work at a discount while you can afford to - flat sharing, living cheaply etc. Learn about both data science AND business. Combining those two skills makes you far more valuable. It also opens more doors.

Find companies you like the look of and pitch them on how you could benefit them without costing them much. From my experience, loads of small companies are always hiring if the right person turns up. Most of the time we're battling to find these people and don't have the money to pay recruiters. They'd far rather take someone on who they can train up at discount because the initial value they add is worth far more than they're able to pay - even if you don't know much. i.e. it's the usual 20% effort = 80% of the value. As the company gets bigger, it's diminishing returns so they have to pay more for less. So to start, it's mutually beneficial if you're both offering each other value at a discount (you skills, them money).

Is this going in the right direction for you? Let me know.
Reply 4
Original post by andrewishiring
I have no experience with the public sector, so all these comments relate to the private sector - specifically startups with between 1-35 people.

You say you've been working for a number of years. What is your concern regarding work load in smaller companies? I would say that if you're looking to learn while changing roles, then you're probably going to have to deal with more than a standard 9-6 job. In most smaller companies there is more work than time simply because the company is trying to figure out how to grow and survive. The people working in them typically love what they do and hence work long hours because it's the only way a startup becomes successful. In larger companies the long hours come from different pressures and the rewards for success are different. This is why someone people love them and some prefer small companies. The work is also different because startups are by definition trying to find market fit rather than executing on a going concern.

The benefit of all this work means that most smaller companies are open to hiring people who may not cost much but who can help them. They will usually tolerate learning on the job if you're not costing them too much - money or time wise. But you'll get exposure to SO MUCH STUFF that if you're s sponge, you'll learn loads and probably open yourself up to other options.

The data science industry is in a massive state of flux as it defines itself. Almost all business can benefit but expert skills are hard to come by - especially while everyone seems to be selling snake oil in terms of skills. This is a double edged sword as some people expect the world while others are understanding of the current state of play and that it's not a form of magic that will solve their problems over night.

Most data science roles are best filled by people who have (or can acquire quickly) a very good understanding of how the business works and what makes it successful. A lot of skills in data science are transferable as long as you understand how to apply them in a different business.

I think you should be able to find some smaller companies who may not be actively looking for someone like yourself but who could benefit from your expertise and desire to learn. If you can afford to, I'd suggest trying to offer your skills at a discount while you skill up. Find an organisation that has a similar set of problems to the ones you've been solving. This doesn't mean the same industry. For instance, maybe you were looking for correlations in dataset of patients at the NHS. This should apply just the same to finding correlations in e-commerce audiences.

I assume you're young. Take advantage of this and work at a discount while you can afford to - flat sharing, living cheaply etc. Learn about both data science AND business. Combining those two skills makes you far more valuable. It also opens more doors.

Find companies you like the look of and pitch them on how you could benefit them without costing them much. From my experience, loads of small companies are always hiring if the right person turns up. Most of the time we're battling to find these people and don't have the money to pay recruiters. They'd far rather take someone on who they can train up at discount because the initial value they add is worth far more than they're able to pay - even if you don't know much. i.e. it's the usual 20% effort = 80% of the value. As the company gets bigger, it's diminishing returns so they have to pay more for less. So to start, it's mutually beneficial if you're both offering each other value at a discount (you skills, them money).

Is this going in the right direction for you? Let me know.



Hi,
Thanks for your detailed response to my questions. By workload I was just concerned about burnout - Ive heard people at startups work 12 hours a day (is this normally at the beginning or even as they grow....obviously work pressures and goals start changing ) . Has job security been a massive issue in startups and do people get laid off a lot ?
So would you suggest just approaching these startup companies directly even if they do not have a job advertised and pitch what I can offer ? The problem being it is difficult to locate them as they normally don't advertise their jobs (from what I understand they recruit through word of mouth or contacts ?). Is there a database of startup companies in different business areas or would you suggest attending networking events etc ?

Money is not an issue for me at the moment - I have been saving wisely for a while unlike a lot of other young people. Although some renumeration would be great, I am really desperate to build up my project portfolio and skill set.
I was considering going back to university and getting a degree in data science but not sure if that would put me at any advantage compared to if I had that work experience and project portfolio under my belt. What is your opinion on this - can you reach the level of a junior data scientist in 6 months by working on data science projects (be it through open source data sets, working in a company, meet ups/free events or local conferences on data science etc) and doing self study through MOOCs ? Would you suggest specialising in one area of data science e.g. machine learning or is a broad experience in other areas also valuable in a job? Sorry for all these questions - I just want to make sure I am thinking on the right track.

Ive seen a lot of jobs being advertised as data science/data analyst/ business analytics - information online seems to suggest positions advertised as 'data science' are more statistical modelling and computationally focussed than those advertised as 'data analyst' positions (which require mainly excel skills) and business analyst are more consultancy/business focussed and look more at risk etc (does actuarial science come under this bracket) ?
Ive read somewhere that it is more advisable to move into a data analyst role and then data science, as its an easier transition - is this true ? The field of big data seems to be growing - I hope it does continue and not fade away in the future. Also, I do hope a lot of new healthcare startups start coming up.
Original post by rk1103
Hi,Thanks for your detailed response to my questions. By workload I was just concerned about burnout - Ive heard people at startups work 12 hours a day (is this normally at the beginning or even as they grow....obviously work pressures and goals start changing ) . Has job security been a massive issue in startups and do people get laid off a lot ?

Let me refer specifically to the London startup scene as that is what I have experience with. The way startups work should be the same elsewhere but the number of startups here gives liquidity to the employment market which is important with regard to what I'm going to say.
I know it's hard to believe, but when you love what you do you'll not even notice the 12 hour days. A 12 hour day in a startup is a totally different experience to one in a big organisation. The very nature of a startup means you're cranking out the long hours and seeing direct results. Often (not always) in a big organisation you're very isolated from the results - so it's harder to feel fulfilled. In a startup you're likely working on a wide range of issues and dealing with people with very wide ranging skill sets. This makes the job, in my opinion, far more interesting. That said, the idea that you do a 12 hour day every day of the week, every week of the year is a fallacy. People work long hours when needed. When you're a junior in a large organisation, you're probably working long hours just because someone told you to. The reasons and motivations are entirely different.
Don't let people exploit you. Move company if they do. Data science is going to be around for a very very long time. I wouldn't think twice about making it a career and betting your future on it. Every single organisation needs these skills and the need for them is spreading rapidly to sectors other than technology as those sectors embrace technology more. So making a jump to another company is easier than you think.
Original post by rk1103

So would you suggest just approaching these startup companies directly even if they do not have a job advertised and pitch what I can offer ? The problem being it is difficult to locate them as they normally don't advertise their jobs (from what I understand they recruit through word of mouth or contacts ?). Is there a database of startup companies in different business areas or would you suggest attending networking events etc ?

You're right about most companies not advertising roles. It's expensive and attracts too many recruiters that waste their time. They'd far rather hire by word of mouth as it filters out he loonies. Find companies that may look interesting by looking at sites that help startups recruit, raise money, etc. Try looking at companies listed on Angel List (they are very respected) https://angel.co/ and companies attending Silicon Milk Roundabout https://www.siliconmilkroundabout.com/ (attend this yourself! it's free)
Remember that for every startup in those lists, there are like 20 that are not. You can often find other startups by looking at the websites that provide services to startups and have a "client list" page. Technical services are a great start. A good example is https://segment.com/ - look at all the companies this company integrates their product with. Get in touch. Tell them what you're about and why you like what they do.
If you manage to get a role somewhere, you'll meet people. Over time they will move on to other companies. Keep in touch. Network like mad. When you need a new role, you can ask around and thanks to companies hiring via word of mouth, you'll be surprised at your options.
This is why I wouldn't worry about being made redundant. It's true, startups are more risky. But you're trying to skill up. It's a trade off. But it's FAR easier to get hired in startup land than big business because there's less red tape and far more pressure on startups to find someone, anyone, who is articulate, passionate and competent, to help. You'd be amazed how hard it is to do this!

Original post by rk1103
Money is not an issue for me at the moment - I have been saving wisely for a while unlike a lot of other young people. Although some renumeration would be great, I am really desperate to build up my project portfolio and skill set.I was considering going back to university and getting a degree in data science but not sure if that would put me at any advantage compared to if I had that work experience and project portfolio under my belt. What is your opinion on this - can you reach the level of a junior data scientist in 6 months by working on data science projects (be it through open source data sets, working in a company, meet ups/free events or local conferences on data science etc) and doing self study through MOOCs ? Would you suggest specialising in one area of data science e.g. machine learning or is a broad experience in other areas also valuable in a job? Sorry for all these questions - I just want to make sure I am thinking on the right track.

It all depends on your actual skill level. If you can code and if you're tech savvy, I'd do online courses and get into a job asap. You'll learn so much so quickly that will give you a solid base that makes you highly desirable to companies. This will give you a grounding that will let you formalise your skills if you wish. If you think your maths isn't up to scratch, I'd say that's the one that's really going to be help by formal education. Especially if you want to do proper data modelling etc.
Don't specialise. The industry is still defining itself. Start broad. I would focus on things that allow you to find trends in data. This is the most usable and most immediately beneficial skill you can bring to a company. Predictive functions etc only start to help once the initial questions have been asked and a meaningful result found. You need to be able to get to this result. That's why understanding what makes a company works is important.
Original post by rk1103
Ive seen a lot of jobs being advertised as data science/data analyst/ business analytics - information online seems to suggest positions advertised as 'data science' are more statistical modelling and computationally focussed than those advertised as 'data analyst' positions (which require mainly excel skills) and business analyst are more consultancy/business focussed and look more at risk etc (does actuarial science come under this bracket) ?Ive read somewhere that it is more advisable to move into a data analyst role and then data science, as its an easier transition - is this true ? The field of big data seems to be growing - I hope it does continue and not fade away in the future. Also, I do hope a lot of new healthcare startups start coming up.

If a recruiter wrote the spec, you're almost certainly getting a biased view of the role. They generally have no clue what they're doing or saying. When I interview anyone via recruiters, it's a crapshoot as to whether the person actually has any of the skills I asked for. If you know this, then you know that the job spec is probably suspect. If the job spec is direct from the company, remember it's probably been written to apply to enough people that the company actually gets replies. There is almost always an open position for someone if they have the right cultural fit and can add some value. This is important to remember as you can often make a position if you understand their needs.
In larger companies the distinction in role is more apparent and more formal. This is why it's harder to get into them when you're starting out.
Another long reply. Probably some unanswered questions but I have to run. Let me know which areas you want more info on. Happy to keep answering as I hope this will help others.
Original post by rk1103

Ive seen a lot of jobs being advertised as data science/data analyst/ business analytics - information online seems to suggest positions advertised as 'data science' are more statistical modelling and computationally focussed than those advertised as 'data analyst' positions (which require mainly excel skills) and business analyst are more consultancy/business focussed and look more at risk etc (does actuarial science come under this bracket) ?
Ive read somewhere that it is more advisable to move into a data analyst role and then data science, as its an easier transition - is this true ? The field of big data seems to be growing - I hope it does continue and not fade away in the future. Also, I do hope a lot of new healthcare startups start coming up.


I think you're pretty much right about data analyst vs data science. The latter is a newer term and has become pretty butchered in meaning - and will probably differ by industry to a degree. The point is that a data analyst is probably more on the business / accounting side than the technology side of things. In a startup the difference in role is more blurred as you're likely straddling both. The atuarial science side is the more traditional side of the analyst type of role. The data science side is newer and often related to big data (another butchered term) and machine learning. In other words, it's probably more related to coding to identify things than Excel - although anyone who is worth their salt knows that Excel is a vital tool in their arsenal. Data science related languages and libraries depend on a knowledge of R / Python etc. Having strong maths background will really help in both roles.
Reply 7
Original post by andrewishiring
Let me refer specifically to the London startup scene as that is what I have experience with. The way startups work should be the same elsewhere but the number of startups here gives liquidity to the employment market which is important with regard to what I'm going to say.
I know it's hard to believe, but when you love what you do you'll not even notice the 12 hour days. A 12 hour day in a startup is a totally different experience to one in a big organisation. The very nature of a startup means you're cranking out the long hours and seeing direct results. Often (not always) in a big organisation you're very isolated from the results - so it's harder to feel fulfilled. In a startup you're likely working on a wide range of issues and dealing with people with very wide ranging skill sets. This makes the job, in my opinion, far more interesting. That said, the idea that you do a 12 hour day every day of the week, every week of the year is a fallacy. People work long hours when needed. When you're a junior in a large organisation, you're probably working long hours just because someone told you to. The reasons and motivations are entirely different.
Don't let people exploit you. Move company if they do. Data science is going to be around for a very very long time. I wouldn't think twice about making it a career and betting your future on it. Every single organisation needs these skills and the need for them is spreading rapidly to sectors other than technology as those sectors embrace technology more. So making a jump to another company is easier than you think.

You're right about most companies not advertising roles. It's expensive and attracts too many recruiters that waste their time. They'd far rather hire by word of mouth as it filters out he loonies. Find companies that may look interesting by looking at sites that help startups recruit, raise money, etc. Try looking at companies listed on Angel List (they are very respected) https://angel.co/ and companies attending Silicon Milk Roundabout https://www.siliconmilkroundabout.com/ (attend this yourself! it's free)
Remember that for every startup in those lists, there are like 20 that are not. You can often find other startups by looking at the websites that provide services to startups and have a "client list" page. Technical services are a great start. A good example is https://segment.com/ - look at all the companies this company integrates their product with. Get in touch. Tell them what you're about and why you like what they do.
If you manage to get a role somewhere, you'll meet people. Over time they will move on to other companies. Keep in touch. Network like mad. When you need a new role, you can ask around and thanks to companies hiring via word of mouth, you'll be surprised at your options.
This is why I wouldn't worry about being made redundant. It's true, startups are more risky. But you're trying to skill up. It's a trade off. But it's FAR easier to get hired in startup land than big business because there's less red tape and far more pressure on startups to find someone, anyone, who is articulate, passionate and competent, to help. You'd be amazed how hard it is to do this!


It all depends on your actual skill level. If you can code and if you're tech savvy, I'd do online courses and get into a job asap. You'll learn so much so quickly that will give you a solid base that makes you highly desirable to companies. This will give you a grounding that will let you formalise your skills if you wish. If you think your maths isn't up to scratch, I'd say that's the one that's really going to be help by formal education. Especially if you want to do proper data modelling etc.
Don't specialise. The industry is still defining itself. Start broad. I would focus on things that allow you to find trends in data. This is the most usable and most immediately beneficial skill you can bring to a company. Predictive functions etc only start to help once the initial questions have been asked and a meaningful result found. You need to be able to get to this result. That's why understanding what makes a company works is important.

If a recruiter wrote the spec, you're almost certainly getting a biased view of the role. They generally have no clue what they're doing or saying. When I interview anyone via recruiters, it's a crapshoot as to whether the person actually has any of the skills I asked for. If you know this, then you know that the job spec is probably suspect. If the job spec is direct from the company, remember it's probably been written to apply to enough people that the company actually gets replies. There is almost always an open position for someone if they have the right cultural fit and can add some value. This is important to remember as you can often make a position if you understand their needs.
In larger companies the distinction in role is more apparent and more formal. This is why it's harder to get into them when you're starting out.
Another long reply. Probably some unanswered questions but I have to run. Let me know which areas you want more info on. Happy to keep answering as I hope this will help others.


Hi,
Thats again for you solid and detailed responses to my questions - it is certainly clearing things up in my mind. Some more questions based on your responses above (generated mainly because of my ignorance of the startup culture/scene) -

1) When you say people work long hours due to the passion - does that mean there is less bureaucracy in a startup compared to larger companies. I have experienced a lot of bullying in my previous workplaces - run by control freaks who try and derail your career by forcing tasks upon you that you are less skilled/trained to do or less interested in doing just because it happens to be stated in small print in your 'job description'. Is there more autonomy in a startup and are managers more understanding that the process takes time or do they expect answers every week.

2) I can code - but can't touch type like a programmer/software engineer. I think like a scientist (well i am a scientist)- breaking down a problem into small parts and testing/re-testing/ validating and getting new ideas in the process. So I wouldn't be good on coding something that requires an answer every day or every week. From what you say - I gather that this is not the case in the startup scene (well I guess its employer dependant). I guess I am going to have to see how it goes in 6 months time.

3) How often do you have meetings - are these normally informal meetings or long formal ones which take up so much time in which you could have completed important tasks ? Is the 'relaxed environment' mentality really true of startups ? I am someone who works best by not having someone constantly spying/micromanaging me 24-7. Do people tend to have a wide range of personalities or is the environment more suited to extroverts (I am more of an introvert )?

4) By recruiter do you mean agency ? I never knew they were the ones who wrote the job description - i always thought they acted as the middle man for every aspect of the recruitment process. This is the response I got from a recruiter/agency when I applied for a data scientist role:

"Thank you for reaching out and applying for my Data scientist role. Unfortunately my clients require candidates who have experience or at least exposure within marketing/media/digital communications. Although you will no doubt have fantastic ability when it comes to working with and modelling data, probably even larger data sets then my clients use. Given your background, we simply wont be able to help you"

So I guess my worry is (based on this response) that I may not be able to move into another business area (in this case marketing) since I have a background in health ? Based on what your response I shouldn't really care about what they think - as you say they seem to go through the skill set as a 'tick box' exercise ?

5) Thanks for links to the website for startup databases and the startup jobs fair. I had heard of silicon milk roundabout but found most of their companies too marketing/social media focussed and jobs too tech/software engineering
Maybe I should still attend one of them anyway. What is the best way to leave a mark on the company representative at such a fair rather than just dropping ones CV on their desk. Are they looking for a project portfolio as well ? What would you be expecting someone approaching you at such a fair to ask you ? Naturally knowing about the company would be good but with so many companies attending it gets quite difficult to remember what each one does (especially if someone is new to the startup scene).

6) What is the interview process normally like - are the questions quite technical or more competency/problem solving based ? What would you suggest is the best way for preparing for them ? Are there cues to look out for/ specific questions to ask which would indicate whether the company environment is friendly/supportive etc . Obviously in larger companies I would be asking about training opportunities/progression/attending conferences etc but not sure if this applies to startups too ?

7) So i take it your a data scientist then ? Hope you don't mind me asking - did you have a maths/science degree and did you do a higher degree in data science as well ? Was it something you went into straight from university or did you discover your passion after a number of different roles doing other things ?

Sorry for such a detailed post again and such a long list of questions - but I guess it also helps others and generates a better awareness of the startup scene as an alternate source of employment rather than going through the typical draconian formal interview process that most larger companies have - which most often than not can put really good candidates at a disadvantage.

Thanks again so much for your detailed responses.
(edited 8 years ago)
Reply 8
Original post by andrewishiring
I think you're pretty much right about data analyst vs data science. The latter is a newer term and has become pretty butchered in meaning - and will probably differ by industry to a degree. The point is that a data analyst is probably more on the business / accounting side than the technology side of things. In a startup the difference in role is more blurred as you're likely straddling both. The atuarial science side is the more traditional side of the analyst type of role. The data science side is newer and often related to big data (another butchered term) and machine learning. In other words, it's probably more related to coding to identify things than Excel - although anyone who is worth their salt knows that Excel is a vital tool in their arsenal. Data science related languages and libraries depend on a knowledge of R / Python etc. Having strong maths background will really help in both roles.


Sorry adding to my post above - by maths skills do you mean possessing an understanding of how the algorithms work ? From my understanding - don't most data scientists use open source libraries which have algorithms already coded in rather than coding from scratch ? Obviously, they would need to modify these to account for their data set/ specific problem they are trying to solve ? A deeper mathematical operations wouldn't be necessary as there are toolboxes/in built functions that one can use ? for example, if I wanted to perform a fourier transform operation, I would just use an in built function although I would understand what it is doing.
Original post by rk1103
1) When you say people work long hours due to the passion - does that mean there is less bureaucracy in a startup compared to larger companies. I have experienced a lot of bullying in my previous workplaces - run by control freaks who try and derail your career by forcing tasks upon you that you are less skilled/trained to do or less interested in doing just because it happens to be stated in small print in your 'job description'. Is there more autonomy in a startup and are managers more understanding that the process takes time or do they expect answers every week.

All companies are different - the same goes for startups. You can work with a bunch of idiots, liars, cheats just as easily as a bunch of people who become life long friends or people who become amazing mentors. The difference in most startups is that there are fewer people to do the work and thus you find that work just has to get done - by anyone. Most of the time there is zero red tape. The red tape usually only comes from legal regulation. In larger companies it comes from that and the need to have standardised ways of doing things and ensuring the correct checks and balances are applied. It's inherent in their nature. In a startup with less than 15 people, you're very unlikely to have more than 3 "managers", and even then, they're probably just the most senior person in their area. Less than 30 people and you probably have some teams with more formal lines of management. But even up to 100 people I've known very flat management styles to be the norm.
Work is usually free flowing. Everyone discusses what they're doing and everyone usually knows what everyone else is working on. So feedback and deadlines are often very informal and set by the people doing the work or as a team. Depending on people's skills, they may or may not have a good understanding of what you're doing and the complexities involved. It's up to you to explain and set targets. It's a team effort. You're ultimately responsible and it's very clear to everyone who is delivering and who isn't. At the same time, people are usually far more forgiving of issues because they usually see them coming and they have the same issues themselves.
It sounds like you've only really experienced very structured environments. This isn't a bad thing. But it's different to a startup. Some people love them and their lack of defined structure and the freedom they afford, others do not.
Now, very very importantly - the entire reason I suggested startups is because it may allow you to work at a discount for someone while you skill up. This is very unlikely to be possible without all the usual formal credentials in a larger org. You may find the more structured approach is better for you - but I'm just trying to expose these other options that many people are not aware of.
Original post by rk1103
2) I can code - but can't touch type like a programmer/software engineer. I think like a scientist (well i am a scientist)- breaking down a problem into small parts and testing/re-testing/ validating and getting new ideas in the process. So I wouldn't be good on coding something that requires an answer every day or every week. From what you say - I gather that this is not the case in the startup scene (well I guess its employer dependant). I guess I am going to have to see how it goes in 6 months time.

In a startup, speed is of the essence - but not in the way you may think. Time is money. And money in a startup is probably one of the most limited resources. What is critical is to know when you're on the wrong path. As they say, you need to "fail fast" - know when to change direction. So people are constantly discussing things and looking for answers. But not like you're probably used to. Obviously you may be working for some maniac, but then you should probably be reconsidering who you're trying to learn from. Remember, don't get exploited. Your sanity and self value are not worth it.
It sounds like you're very analytical. This is good. The more you practice coding and solving problems the better you'll get. Maybe getting better at touch typing would be something to focus on - it shouldn't take more than a few weeks of 20 mins a night practice to really get to grips with it.
Startups are about constant discussion and feedback with the rest of the team. Just like any other team - except you get a much wider view of what's going on in the entire company and the issues they're facing / how they're benefiting from what you're doing.
Original post by rk1103
3) How often do you have meetings - are these normally informal meetings or long formal ones which take up so much time in which you could have completed important tasks ? Is the 'relaxed environment' mentality really true of startups ? I am someone who works best by not having someone constantly spying/micromanaging me 24-7. Do people tend to have a wide range of personalities or is the environment more suited to extroverts (I am more of an introvert )?

Startups generally mean small teams who are aggressively focussed on trying to find market fit for their product so that they can scale. Their culture can vary. It all comes down to who founded it and the people they surround themselves with. The majority I know / have founded / worked with have been exactly the kind of relaxed culture you're thinking of. Work hard, party hard (when I was younger), celebrate the achievements, celebrate the failures (you learned something at least!) and generally try to make an environment people want to be in every day. Being able to talk to people is critical - maybe a bit more so - in a startup because everyone needs to be on top of things. That said, because you all talk the whole time, I think it works better for those are more introverted because they get to know all the people and work with them all the time. You're not constantly having to deal with new people who people you only sort of know. Meeting wise, there's generally a daily stand up culture and then weekly/monthly meetings to check roadmaps etc. But every company is different.
Original post by rk1103
4) By recruiter do you mean agency ? I never knew they were the ones who wrote the job description - i always thought they acted as the middle man for every aspect of the recruitment process. This is the response I got from a recruiter/agency when I applied for a data scientist role:
"Thank you for reaching out and applying for my Data scientist role. Unfortunately my clients require candidates who have experience or at least exposure within marketing/media/digital communications. Although you will no doubt have fantastic ability when it comes to working with and modelling data, probably even larger data sets then my clients use. Given your background, we simply wont be able to help you"
So I guess my worry is (based on this response) that I may not be able to move into another business area (in this case marketing) since I have a background in health ? Based on what your response I shouldn't really care about what they think - as you say they seem to go through the skill set as a 'tick box' exercise ?

Some recruiters will write specs in an attempt to lure candidates even if they don't have roles. They may also alter specs because they're getting no bites.
That response sounds pretty typical. If they're asked to fill a role and they've got candidates who have more experience, you're going to have a very tough time. This is the whole point of trying to get yourself a role without applying for one that exists. You want to get one on the basis that you like the company, they like you, they need your skills and can afford those skills - hence working at a discount for a bit while you skill up.
Recruiters for the most part will just tick boxes. In my experience, a good hiring pipeline at a company looking to fill a single tech roll with a good quality candidate will have, on average, 20 candidates, 10 of which will get a face to face interviews. You can see now why recruiters take anything they can get. It's a numbers game. This is also why hiring is such a time suck and why many companies just don't have the time to be actively hiring but will hire the candidate out of the blue because they fit.
Original post by rk1103
5) Thanks for links to the website for startup databases and the startup jobs fair. I had heard of silicon milk roundabout but found most of their companies too marketing/social media focussed and jobs too tech/software engineeringMaybe I should still attend one of them anyway. What is the best way to leave a mark on the company representative at such a fair rather than just dropping ones CV on their desk. Are they looking for a project portfolio as well ? What would you be expecting someone approaching you at such a fair to ask you ? Naturally knowing about the company would be good but with so many companies attending it gets quite difficult to remember what each one does (especially if someone is new to the startup scene).

Attend them - if only to practice meeting people in a setting where understanding each other's needs is highly time boxed. Remember, that these guys often talk to each other. You may not get a role with one of them there, but they may suggest you talk to someone else. You also meet other candidates. Some times i think those connections become even more valuable. The companies pitching there feeling just as awkward. :smile: When I've done these, most candidates like to ask why they should work for you. What you're doing and why it's important / interesting. It's often a reverse interview where the company is pitching the candidates. Most of the time it's 2-5 mins of chat and then maybe leaving details / CV and then following up. Silicon Milk Roundabout will give you a booklet telling you about each company, so remembering it's too hard.
Your industry is probably under represented, but I'd still go along for the reasons I mention above. Remember these companies pay £££ to attend. So there are many that can't afford to. You probably want to be targeting those. You just have to find them.
Original post by rk1103
6) What is the interview process normally like - are the questions quite technical or more competency/problem solving based ? What would you suggest is the best way for preparing for them ? Are there cues to look out for/ specific questions to ask which would indicate whether the company environment is friendly/supportive etc . Obviously in larger companies I would be asking about training opportunities/progression/attending conferences etc but not sure if this applies to startups too ?

Interviews are different everywhere. Some people are terrible at doing them. Some are good. Don't try to game them. Do this... Know your subject matter. Know how it applies to the business. Do research on what they do and what their likely issues are that they're facing. Then, try to get the interviewer to have a discussion where you talk through some of their challenges and how the TWO of you may go about solving them, while you try to weave your knowledge and any experience into the solution. If the person doing the interview has any skills they'll do this because it's the fastest way to work out what it's going to be like working together solving problems. Most of all, try to have a discussion. Be afraid of anywhere that smashes you with questions / tests.
Original post by rk1103
7) So i take it your a data scientist then ? Hope you don't mind me asking - did you have a maths/science degree and did you do a higher degree in data science as well ? Was it something you went into straight from university or did you discover your passion after a number of different roles doing other things ?

I'm not actually - I'm a CTO and developer/system architect by trade. I have nothing more than a degree in information systems. Everything I know is pretty much self taught or through working with great developers and business people. I was lucky that my passion for the internet lead to my initial job which gave me time to skill up programming. I finished my studies and just kept taking roles that interested me as people I knew moved companies and I sometimes followed.
The good data science people I know have two things in common. They enjoy solving problems and enjoy solving them with code and maths. Some have come from formal math backgrounds others from formal comp sci backgrounds. The key is that the stuff that's made them great has pretty much all been self taught while trying to solve real world problems. With the advent of all the tech stacks and libraries, the application of these existing tools to solve problems makes the domain much more accessible. But having an understanding of how they work under the hood gives you a big advantage when it comes to optimising and tailoring. But, 80% of the value comes from 20% of the effort. The 80:20 rule always applies.
Original post by rk1103
Sorry for such a detailed post again and such a long list of questions - but I guess it also helps others and generates a better awareness of the startup scene as an alternate source of employment rather than going through the typical draconian formal interview process that most larger companies have - which most often than not can put really good candidates at a disadvantage.
Thanks again so much for your detailed responses.

Hiring is such a crap process. It's a wonder anyone gets hired for the right role. In most cases it's about tick boxes and how much chemistry there is between a few people for your 60 minutes of interview. Would you choose a husband/wife in so little time? This is why word of mouth hiring is so beneficial and why networking and just meeting others who can intro you is so important.As someone who is in the startup world trying to build teams, I wish more people knew about our world as it would significantly widen the pool of talent we're looking to hire from. I've had many people come direct from studying some odd degree, skill up with us in some digital related role they had no idea about and be very happy and challenged by the industry they've landed in.
Original post by rk1103
Sorry adding to my post above - by maths skills do you mean possessing an understanding of how the algorithms work ? From my understanding - don't most data scientists use open source libraries which have algorithms already coded in rather than coding from scratch ? Obviously, they would need to modify these to account for their data set/ specific problem they are trying to solve ? A deeper mathematical operations wouldn't be necessary as there are toolboxes/in built functions that one can use ? for example, if I wanted to perform a fourier transform operation, I would just use an in built function although I would understand what it is doing.


sorry, should have covered this in the other reply. As I said there, having an understanding of the inner working of these libs is highly beneficial when fine turing or working out why certain things aren't working or would comparing performance. but getting going, sure, just being able to apply these to the business case is the most critical part and the one that will get that 80% value with 20% effort. This is where you should focus your efforts when starting somewhere. You want to be able to give insight that the other people can't get because they either don't have the time or the skills to even apply these tools to the data at their disposal. But asking the right questions of the data is key. This is harder than it sounds and is why you must get to grips with what they're doing and how their data is produced before diving in and promising the world :smile:

=====

I really hope these answers have given you and others some insight into startups and data science. :smile:)
Reply 11
Original post by andrewishiring
sorry, should have covered this in the other reply. As I said there, having an understanding of the inner working of these libs is highly beneficial when fine turing or working out why certain things aren't working or would comparing performance. but getting going, sure, just being able to apply these to the business case is the most critical part and the one that will get that 80% value with 20% effort. This is where you should focus your efforts when starting somewhere. You want to be able to give insight that the other people can't get because they either don't have the time or the skills to even apply these tools to the data at their disposal. But asking the right questions of the data is key. This is harder than it sounds and is why you must get to grips with what they're doing and how their data is produced before diving in and promising the world :smile:

=====

I really hope these answers have given you and others some insight into startups and data science. :smile:)


Thanks very much for your very detailed responses to all my questions - things are a lot clearer now. Hopefully, I end up working in a startup soon - probably even playing a bit of ping ping, enjoying free food and sliding between floors after a hard days work
interesting post sorry to impose but if any experienced data analysts are about i would like to have a private chat about a project that i am working on and would appreciate any help
Original post by Singh89
Hello TSR

I would like know about anyone's experience working as Data Analyst or in the area of Data Analytics. I have 1st class degree in chemistry from RG university with a good grounding with SQL server/scripting.

Look forward to your responses.

Cheers.


Firstly, What type of company for work you are looking for? There will be a difference between a little startup and big company. To make a long story short, I have attached the main skills Data Analyst should have. Also more info you can find here

Original post by Singh89
Hello TSR

I would like know about anyone's experience working as Data Analyst or in the area of Data Analytics. I have 1st class degree in chemistry from RG university with a good grounding with SQL server/scripting.

Look forward to your responses.

Cheers.


I have worked as a data analyst for the past few years. The job can be extremely tedious and time-consuming, but it can also be very rewarding. As a data analyst, you are responsible for sorting through large amounts of data and extracting relevant information. This information is then used to help make business decisions or to improve business operations. I found the job using this company that offers a list of jobs data analysis opportunities from different companies called data jobs

Latest