WE SPECIALISE IN FINDING FANTASTIC OPPORTUNITIES
FOR DIGITAL AND DATA SPECIALISTS WITH THE MOST INNOVATIVE BUSINESS ACROSS EUROPE AND THE USA.
[WATCH] Replay of our Machine Learning and Finance Meetup
[WATCH] Replay of our Industrial Application Data Science Meetup
Darwin Recruitment returned to Copenhagen for its second Data Science MeetUp. This time, we had three speakers talking about the industrial application of Data Science. Our Data Science Specialist Luke Parkinson and our Team Leader for Denmark were in attendance with more than 50 Data professionals in the audience. Speakers and Topics Dmytro Iakubovskyi - Postdoctoral Data Scientist and Machine Learning Engineer – Discovery Center, Copenhagen University Academic researchers are thought to be well apart from the industry. But there is also a constant demand for academic researchers in rapidly evolving industries, such as Data Science. For such a transition, the majority of recommendations is concentrated on new skills that need to be gained by an academic researcher. In this talk, I will focus on details on specific capabilities of academic researchers that are in high demand in industry-based Data Science. Jakob Rasmussen – CEO - Apex AI The video game industry has been late in adopting machine learning. This is now changing, with many new initiatives such as EA Games' SEED lab and Unity incorporating AI into the core game engine. AI might thus transform the video game industry in ways it is transforming other industries. However, AI for the video game industry can not necessarily use the same approaches as AI for other industries, as some of the challenges to good AI in video games are super complex. To address these challenges, I will in this talk present some of the approaches Apex AI has used for machine learning in the upcoming title Unleash, and what we have learned from that process about developing AI for video games. Neil Hodgson – Head of Asset Science – Maersk Digital Every year Maersk transports over 1 million refrigerated containers carrying everything from fruit to pharmaceuticals in controlled temperature and atmospheric conditions. Each container reports back its status every hour, giving us a large amount of data about the journey. In this talk, I'll show you how data science is used every step of the way as your bananas make their way from South America to the shelves of your local supermarket in Denmark.
[WATCH] Replay of our AI in Data Science Meetup
RecdoTech returned to Berlin for their third Meetup in the AI in Data Science series late in February. The night was hosted by our Data Science Specialists Mark O'Toole and Daniel Neaves and was yet another great success. Over 65 professionals attended and more logged online to listen to guest speakers share their experiences and insights. Talks included: What is the big in big data? By Dat Tran, Head of Data Science at Idealo Internet GmbH. Dat shared his experiences from building up a data science team at the largest comparison service in Germany. Noa Tamir, Data Science Team Lead at Babbel, presented some strategies for creating a data-driven culture in your company without teaching everyone statistics and algorithms. Colin Clark, a technology specialist in the field of Big Data at CoinBau, discussed various areas required to effectively adopt data science, machine learning and AI. And Maria Borner, AI Engineer at Xain, shared the details of her current work project that is implementing AI for a completely automated accounting system. We run our AI in Data Science Berlin series of Meetups on a bi-monthly basis, so check out our events page for our upcoming events.
[WATCH] Replay of Blockchain Meetup
Our recent Blockchain Meetup, hosted by our Executive Frontend Specialist Jodi Barrow (Jodi.Barrow@darwinrecruitment.com), was a massive success. Over 75 tech professionals attended to listen to our experienced and prestigious speakers deliver insightful and informative talks. We plan to host another Blockchain Meetup in Berlin to capture the rise of the Blockchain and continue to give back to the professionals and community. Head over to our events page to see all our upcoming events. As specialist recruiters, we can add value to businesses with strategic recommendations and market insights. Start by downloading our North Germany Market update containing salary information, market analysis and market summary for each technology we recruit in here. You can also sign up for our newsletter to keep up to date with the latest events, industry news and career advice by following this link.
[WATCH] Replay of our Attracting & Retaining Women in Tech Meetup
[WATCH] 10 Key Points from our first AI in Data Science Meetup
We finally did it! Hosting a meetup is something that myself and my team had been wanting to do for a while, but never actually got around to organising one. We had an amazing response, with 153 people signing up for the event - filling the venue to 90% capacity! Pretty good, right? By the way, if you're looking for a Data Science role in Berlin, I probably have the perfect job for you. You can also see how much you could earn in Berlin as a data scientist by downloading our Market Update. We hosted our AI in Data Science meetup at Fusion Factory, right in the heart of Berlin: Our second speaker was Sebastian Foucaud, who spoke about the Democratization of Machine Learning Products and the Impact on Business. These are some of the points from his presentation that I found interesting: 1. Artificial Intelligence is not SciFi - it is HERE and NOW (6:13 in the video) This is something I already knew, but having Sebastian talk about it really cemented it for me. He said it's so important to technology businesses, that if the company hasn't got any views on it, then you are already too late. We have it in our phones, homes, cars and even the grid power! That just goes to show how critical it is already. Imagine how vital it will be in five years? 2. We can build any AI-based product, but it doesn't make it a good idea (7:45 in the video) The trend for consumer technology products at the moment is to advertise the AI & machine learning (it's almost become a buzzword at this point). But does it make it better? Or useful? Is it even necessary? "AI for AI's sake is never the right way" - Sebastian Foucard Simply stated: 3. The model doesn't matter, it's the execution! (11:30 in the video) This just goes to show that a data scientist really can work with what they are given (most of the time). Some ways might be easier, some might make it more difficult. But: If you have the right technology and talents, it should work! And there are tons of open-source software available, and there is a huge amount of talented data scientists. I only work the Berlin market and have connections with thousands of them! My experience is that: 4. You need to know your customer before they visit your website (or app) (24:56 in the video) User experience is vital for any business, app or website. It can be the difference between a paying and returning customer, to a customer who never interacts with your brand again. Sebastian made a great point about when he was looking to buy a car. The app wanted to know what make of car he would like to buy. He had no idea (and he didn't care). He wasn't looking for a particular make of car, he was looking for a car that would suit his needs. So, that means: 5. Personalised recommendations are the way to go (28:28 in the video) The last point which stood out to me about Sebastian's presentation was how he maximised the profits from users who used two different platforms. There are two different platforms, one for buying houses, and one for buying cars (only 5% of users are on both). Somebody viewing a farmhouse was shown an advert for a truck. See the connection? Information was plugged in from other users to achieve this recommendation. These are just a few of the interesting points from Sebastian's presentation, you can see the rest by watching the full video here: Our other speaker on the night was Luba Weissman, who spoke about the Potential of Behavioral Data in Credit Scoring. Luba has been a data scientist for the last 10 years now, and for a big part of her career, she's been in the finance industry. Here are 5 points from her presentation that I found interesting: 1. AI considers what linear models could not (3:28 in the video) Linear credit scoring models in the past have been limited to what they can analyse and make decisions on. This affects people from all walks of life and around the world! As Luba explains, people without payments/credit history, immigrants and the younger generations would have been denied or not even considered before. AI can look past these factors and delve deeper into making the correct decision. As if that's not enough: 2. Regulators can slow down these processes (4:55 in the video) I suppose this is the same in any industry, really. But this is especially important as the data of huge numbers of people are at risk here. But that's just part of the story: 3. Reality isn't linear (5:36 in the video) This one makes a lot of sense to me. Humans aren't predictable, and regular techniques hit a limit on how often they are right or accurate. AI tries to alleviate this by using non-linear techniques. Think about it: 4. Customer behaviour isn't only limited to Facebook (6:58 in the video) I knew that customer spending behaviour was tracked for credit scoring, but I had no idea just how much was tracked. Using your card in a casino (or at an ATM close by) has an impact. Change your smartphone? It'll have an effect. Call customer service a lot? That'll have an effect too. Crazy stuff, right? Even worse: 5. We GIVE companies this data As soon as I heard what has an impact on creditworthiness, my next logical question was. How do they get it? I'm sure yours was too. Turns out, we give it to them. When we sign up on websites or use the apps on our phone. Social networks have a big part to play as well. Luba's presentation had a lot of interesting insights into credit scoring and what part AI has to play in it. You can watch the full presentation here:
When Will Artificial Intelligence (AI) Replace Web Development & Frontend Developers?
Artificial intelligence (AI) has made a huge impact in web development, frontend developer jobs and artificial intelligence jobs. This is not a case of artificial intelligence vs web development; they have to work together. Artificial intelligence in web development/web design is becoming more and more common and the impact of artificial intelligence on web development is clear. We can see the rise in artificial intelligence popularity grow rapidly here: We held a webinar about this exact topic with Francisco Ramos, Expert Frontend UI Developer at Move Digital AG, and Charles Ahmadzadeh, Co-founder of Bunch.ai. They both gave us their views on if artificial intelligence can design websites and how it is going to fit in frontend and web development (you can watch that webinar here or watch directly below). By the end of this article, you'll know the benefits of artificial intelligence (AI) in frontend development, where it's already being implemented and what the future holds for the technology. Keeping ahead of the curve in the technology industry allows you to better serve your customers, clients or employer. Lucky for you, that's exactly what I'm going to share with you in this article: When will artificial intelligence (AI) replace web development & frontend developers? Let's get into it: Artificial intelligence (AI) is already assisting with the design of websites. Before we get too far into this, I think it's important to say: The most important thing to consider for a website is how easy can the user find the information they're looking for. It's as simple as that. Websites have become more and more complicated as times goes on, but the most important factor is still the user experience. Artificial intelligence (AI) should be used to enhance the user experience or reduce the frontend developers workload, not complicated things. Frontend development is advancing at a quick rate and artificial intelligence (AI) is helping fuel this, especially with making developers more productive. For example: Have you heard of Sketch2Code? This tool from Microsoft claims to "Transform any hands-drawn design into a HTML code with AI." Pretty cool, right? Sketch2Code detects design patterns, understands written text, understands the structure then builds HTML. You can see something very similar in action here: The difference in the GIF above is that this is happening LIVE! I'm sure I don't need to tell you how this could potentially save a lot of time for developers. It could also open up the door for more design-focused people to get involved with web development early on in the process. Does Sketch2Code and other similar tools mean that artists and design-focused individuals are going to start building/designing websites more and more? User experience needs to be considered of course, and there are certain aspects of websites that users have come to expect, but it's still interesting to think about. Artists and designers will certainly have a different idea on how a website should look, which might be perfect for fashion or design brands who are trying to stand out against the competition. Looking at languages and programs that are falling out of favour is an interesting way to see where the technology industry is heading. We decided to take a deep dive into PHP and explore the future of it. You can read that article here or watch the full webinar here. But, back to artificial intelligence, did you know: Artificial intelligence (AI) can help detect bugs too. A group from Microsoft Research, along with Cambridge University, set out to create a model that would detect bugs that no compiler or inter-unit test would be able to find. The DeepCoder is a data generation strategy which models the input-output sets and algorithms for searching over program space - the model is able to write code and learn from a small description of the problem. Here's how it works: You feed the machine a small description of the problem and it will write a few lines of code and try to solve said problem. Though there are some big limitations with this process: the machine can only write in lines of code. Also, there are other experiments that all come with huge limitations, which are not reliable enough to write production work. Pretty interesting stuff, and the model will only continue to improve with huge corporations like Microsoft and Cambridge University working on it. Did you know artificial intelligence plays a major part in healthcare too? Teamed up with blockchain it can be a formidable force. We take a look at this topic in more depth here. Artificial intelligence isn't only limited to web development though: Artificial intelligence is an important part of user experience as well. Let's not forget that artificial intelligence can also be used to enhance the user experience. I'm sure you've all seen one of these pop up at some point in the past week? Chatbots have been integrated into websites for a while now to simplify peoples experience on websites. They can direct people to different areas of the website, answer FAQ's and connect you to the employee who can best help you. A great example of this is Roof AI: The user has received a quick and easy service, and the employee has been sent a lead directly with information attached. What more could you want? Artificial intelligence aiding user experience has a lot of room for growth as well, which we'll talk a little bit more about later on. But on the other hand: AI cannot replace frontend developers in specific tasks. The main question here is whether artificial intelligence could replace frontend developers, which is really asking whether AI could write code. There are a few examples which help put this idea into context: Andrej Karpathy, the director of artificial intelligence and Autopilot Vision at Tesla trained a multi-layer Recurrent Neural Network (RNN). Andrej used an entire Linux repo on Github and compacted the data onto a single, giant file – 500MB of C code to be specific. He then ran a few key functions. With everything in place, including the variable parameter, conditional loop, property indentation, the model functioned adequately. Although he found some mistakes and it didn’t do anything useful, the code did look quite good Pretty good start. Francisco Ramos elaborates on the limitations of the model above created by Andrej Karpathy: "Software developing requires a good understanding of the problem and the business; essentially it requires intuition. We know machines are very good at finding patterns and humans have no chance when competing against machines in infrequent, high-volume tasks but when it comes to solving a problem it hasn’t seen before; machines don’t perform well." Francisco doesn’t think that machines will be able to develop intuition and the ability to interpret business values or features. "I'm certain that machines will evolve to write code one day, although I do not know how reliable this will be." says Francisco. It’s important to not forget that machines will take our code as a reference and will help us along with our workflow. Which means developers will always be needed for that initial input. This further strengthens the idea that AI is able to enhance frontend development, rather than replace it. So, the question remains: What does this mean for frontend developers? It wouldn’t be a bad idea to actually learn about artificial intelligence and machine learning. Francisco says that he has used the framework Tensorflow.js in a few personal projects already. "Tensorflow.js can be used to import models and build and train outside browsers; but it’s also used in building and training, right in the browser using the user GPU." says Francisco. Why would you need machine learning models in the browser instead of in the server-side? "It is GDPR friendly as data never needs the browser; there is no need to send any data to the server (providing better user experience)." says Francisco. Imagine the browser has access to all kinds of sensors today on user input, this would be a great source of user input to play with. You could basically interact with a website by just gesturing, thanks to the library. With Tensorflow.js projects, you don’t need to install a heavy libraries framework, compiler, and interpreters. It’s all in the browser. Here's the deal: Frontend developers should keep an eye out on how the latest technologies are evolving; it is part of the job to be up to date with the evolving technologies - you’ll be replaced quickly if you don’t. The pace it can happen at can sometimes be scary. But you're probably wondering: What is the future of artificial intelligence, frontend and web development? Charles told us "In a couple of years, websites will become even more adaptable and will be able to understand the personalities and emotions of their users when they interact." The website will eventually be able to adapt to these emotions to deliver either a better experience, but also from a business perspective, this will allow for an optimised conversion rate through different funnels. This is crazy. It's also worth noting that a lot of people believe artificial intelligence is going to actually create more jobs instead of replacing humans: In the end, if a company is using certain technology, there should be a reasoning behind it; it shouldn’t just be because artificial intelligence is impressive. This is one of the paths where technology will go, in providing a clear route on investment for a company that would deploy this in their frontend. However, there is a problem that is very frequent and has been yet to be solved. This is the problem of software estimates in predicting bugs and risks. Machine learning could actually be very useful for this problem, in not necessarily replacing predictions but by providing additional data points so that software engineers could make more informed decisions to estimate a user story. If we were able to, we could map out how much code would be needed to be built for a user story and how we’d understand this in the context of the company or the team building the product. Again, this is about tailoring and customising the experience to the user who is using the algorithm in this context. This could actually help developers better estimate, and going beyond estimates, helping with their ability to predict the impact on the maintainability of the codes. I’ve seen a lot of responses saying that artificial intelligence will never replace frontend development. However, there are lots of ideas on where it could go in the future to help frontend developers. Personally, I don’t see a future where developers won’t have a job, but I do see a future where AI will allow them to create better user experiences. In the future, AI will allow developers to have more time to explore other parts in their development process. A company hiring for an artificial intelligence position usually face 4 BIG problems, which I talk more about in this article. Often time is wasted, they're looking in the wrong places or there are issues with ethics. WATCH THE FULL WEBINAR HERE:
[WATCH] 15 Key Points From Our Data Science Meetup in Copenhagen
We recently held our first-ever data science Meetup. *SPOILER ALERT* It was a massive success! Over 100 data professionals attended to listen to our experienced and prestigious speakers deliver insightful and informative talks. That's not all: We plan to make this into a series of talks, to capture the data science community, and to continue to give back to the professionals and startups in Denmark. Head over to our events page to see all our upcoming events. Or, you can see how much you could earn as a Data Scientist in Denmark, by downloading our Market Update. We go through how long it takes to get hired, where the interviewees were located and their language skills, and the average salary rates. Everything you need to start looking at a new position! Anyway. We hosted the Meetup at Founder House, right in the heart of Copenhagen: Our first speaker, George Zira centred his presentation on natural language processing with deep learning. George’s work at TripAdvisor is focal to their application of machine learning to a wide variety of industry systems such as recommender systems, fraud detection, user personalisation and online advertising. Below are some points I found interesting from his presentation: 1. The Problem Of Circumvention (6:42 in the video) With TripAdvisor’s wide variety of industry systems, the problem of circumvention proves to be very challenging. The platform of this business ensures safety for its customers; all communication should happen through the platform to decrease the risk of circumvention and fraud. Bag of words a problem solver? A classified bag of words and regular expressions as features will reduce false positives but as it cannot learn new patterns, the reliance on regular expressions becomes costly. But here's the answer: 2. Deep Learning for the better. (14:48 in the video) There are no more regular expressions. Algorithms can now build their own features! It gets better: More circumvention will be caught. Which will result in accuracy improvement. What more could you want? 3. Model and Training (18:07 in the video) With the model combining recurrent networks and layers, the probability results of circumvention prove natural language processing with deep learning works. Results: 0.998, 0.710, 0.97, 0.99 The model deployment sees a distribution of messages as a function of the probability convention given by the model. You might be wondering: 4. What's The Impact? (20:26 in the video) The feedback from the community after the model’s release saw positive effects: “I thought I was about as sneaky as sneaky could get, but even I can’t get by the filter process at TripAdvisor these days.” “I would agree, the spyware has become more sophisticated and phrases one could get away with in the past can no longer get through.” These are just a few of the comments from users. By the way: 5. “It is no longer black magic” – George Zira (22.02 in the video) It's both easy to set up and easy to work. A flexible tool that proves to be a positive function that works and justifies its place in the market. Ultimately becoming “an important asset of a data scientist.” These are just a few of the interesting points from George's presentation. You can watch the full video at the bottom of this blog post. We also had Nicolas Horst, who is the Head of Devoteam Discovery at Devoteam, come to speak to us about data science as a service. Nicolas spoke about how businesses can get started on their “big data journey” and how data in which they collect can be turned into revenue. Below are 5 points of interest from his presentation: 1. Experts - the most wanted resource. (32:55 in the video) Data has become the ‘new fuel’ in the development and operation of businesses. Becoming a data-driven business, however, proves to be a challenge. When it is difficult to reach critical mass to gain business impact, it is difficult to attract, retain and develop data science experts that provide this new fuel. Where is the core business problem situated? How can value be created? 2. The art of turning data into actions. (40:05 in the video) Teamwork between humans and computers can transform data into insights. The creation of data products equates to the creation of actionable information and a change from traditional analysis approaches. “When you do data products, you have an opportunity to automate the manual process in business” (41.06) Good news for the people reading this article: 3. The data science service is a must-have! (41:38 in the video) Nicolas makes the point that the answer for a business to become data-driven lies in the works of a data science service. The data science service provides: a team of experienced data experts. a secure cloud-based data and analytics platform. readymade data sources for public basic data. predefined methods, models and algorithms. What this means is that they are: 4. “The experts in revealing the value of your data” (41:38 in the video) As the customers are provided with a fully functional data science department, the previous difficulties are overcome. Without upfront investments, customers are ready to start working with business problems. The impact of data science proves to be essential for it has become a “necessity to compete in a data-driven world.” (48.46) Want to know the best part? 5. Data science is the connection. (50:02 in the video) When inductive reasoning is provided, new analytic paths are created. Models are constantly tested, updated and improved until better ones are found. Nicolas emphasises the contemporary importance of data science as it affects aspects of our daily lives such as our phones, travel, even the process of making our doctors’ appointments. Data connected to business allows for revenue to be a well-welcomed outcome. “It is not rocket science…it’s something better” – Nicolas Horst Our third speaker, Wolf Rittershofer is a Senior Data Scientist at Danske Bank. Wolf spoke about working on the end to end implementation of machine learning models, including specification with the corresponding business unit and monitoring business impact. Below are some points of interest from his presentation: 1. Danske Bank’s Centre of Excellence Approach (59.45 in the video) Two centres of excellence are the focal point of this business’ dedicated data science teams; customer analytics and risk analytics. Customer analytics sees push turn into pull and 60 models put into production. Success rate: 60% incomplete pilots, 40% in production creating value, 10-30% industry standard It isn't always simple though: 2. There are challenges working with organisation and clients on the business side. (1.03.00 in the video) The people factor – create business ownership and relate results to what people know. GiGio vs Pizza Balance – good foundation “bad data science is still pretty good data science”. Move fast, improve later. Leverage existing platforms. (1.05.41) Adapt your evaluation metrics to your business case. Data science is becoming more and more integrated as time goes on: 3. Data science supporting the business case. By creating ownership, it provides the right plots and allows businesses to make their own decisions. Most businesses only see the return on investment. Which is still very important, but not the only factor. Precision does not depend on the true negatives. This is very important: 4. “Know your metrics and its pitfalls!” (1.21.05 in the video) Picking the right metrics is essential for a positive outcome. Successful data science and production can only happen when the people factor is involved. What do they want? What do they need? Spend time with the business! The important thing to remember is: 5. You are the new guy on the block. (1.21.52 in the video) Accept that you are the new guy on the block. Speak in terms that business understands and deliver in terms that business expects. Choose your metric wisely, with the right business case you can get ownership and buy-in in from the business. “Always work for model improvement.” – Wolf Rittershofer As specialist recruiters, we can add value to businesses with strategic recommendations and market insights. Start by downloading our Danish Market Update containing salary information, market analysis and market summary for each technology we recruit in here. You can also signup to our mailing list to keep up to date with latest events, industry news and career advice by completing following this link. Watch the full replay of our data science meetup below:
Hosting an event is the best way to bring targeted talent pools directly to your office.
Hosting an event isn't as complicated as it may sound - we help you find suitable speakers, discuss topic ideas, create all marketing collateral and promote the event to your desired talent pool (we also help find a venue if your office isn't suitable).
Interested in hosting an event to boost your recruitment efforts? Fill out the form and we can discuss the best options for you.
WE HELP BUILD BUSINESSES. WE DO THIS BY FINDING THE BEST PEOPLE SO THAT CUTTING EDGE TECHNOLOGY COMPANIES AROUND THE GLOBE CAN GROW, EVOLVE AND CONTINUE CHANGING THE WORLD AS WE KNOW IT.