Appsilon | R Shiny | Machine Learning

BACKGROUND

Please describe your company and your position there.

My company is a leading growth private equity fund. I am a data scientist in the Strategic Resource Group, a team providing support on strategic initiatives to our portfolio companies.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire Appsilon?

Appsilon helped me improve an R Shiny app I had developed for a portfolio company, helping automate it's deployment, improve the architecture and recommended ways of changing the styling.

SOLUTION

How did you select this vendor and what were the deciding factors?

I came across Appsilon through looking at some of the R packages they open source. I then spoke to Damian Rodziewicz at Appsilon who took the time to really understand my role and use-cases for ther help

Describe the project in detail and walk through the stages of the project.

A large part of this project was in helping me understand best practices for Shiny app development. We split the work into two. Firstly we focussed on deployment, where an engineer containerized my application and automated the deployment to Azure including adding single sign-on through Azure Active Directory. The second part was around architecture and styling. A developer reviewed my code and introduced me to ways of structuring Shiny apps, strategies for loading and passing common data through a user's session and tips on layout and styling.

How many resources from the vendor's team worked with you, and what were their positions?

Damian was my initial contact and helped kick off the project Anna helped structure the project by listening (patiently!) to what I was trying to achieve and planning how best to spend our time. Oleksandr worked on the deployment side and took me through the scripts he had developed step-by-step to help me understand how they worked. Pedro introduced me to several best practices for structuring more complicated Shiny apps leveraging different packages and data types. He also recommended several ways of improving and customizing the styling of my apps

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

I have some reusable scripts to use on future projects that will greatly reduce the deployment time, as well as multiple examples of better ways of writing and structuring Shiny apps.

How effective was the workflow between your team and theirs?

Their workflow was very easy and efficient.

What did you find most impressive or unique about this company?

They were very flexible given the relatively loose specifications I had for the project and understood a key outcome was to gain knowledge on best practices given the expertise they hold in the field

Are there any areas for improvement or something they could have done differently?

Not really, it was very useful and a good indicator of how future engagements with them would run

BACKGROUND

Please describe your company and your position there.

I am a founder of a successful software business volunteering as CTO of Lasy i Obywatele (Forests & Citizens), a civic initiative promoting data-driven policymaking in forestry and applying CivicTech to advance transparency and citizen participation in relation to management of public forests.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire Appsilon?

I act as a CTO for a civic initiative Lasy i Obywatele (Polish for “Forests & Citizens”), a group of volunteers who applies CivicTech to environmental advocacy. The initiative has published a map of logging plans in public forests in Poland. I was looking for a feasibility study on the integration of specific historical aerial imagery into this online map.

SOLUTION

How did you select this vendor and what were the deciding factors?

I was referred to Appsilon by a friend when looking for expertise in aerial imagery processing. Positive references from my professional network was the deciding factor for me when I was considering whether to engage with Appsilon.

Describe the project in detail and walk through the stages of the project.

State-owned forests constitute almost one third of Poland's territory. The state forests administration (Lasy Państwowe), a powerful central institution responsible for forestry, is criticised for lack of accountability, poor transparency and short-sighted management of high-value forests (e.g. Puszcza Białowieska primary forest). Lasy i Obywatele (LiO, Polish for “Forests & Citizens”) is a civic initiative promoting data-driven policymaking in forestry and applying CivicTech to advance transparency and citizen participation in relation to management of public forests.

LiO has published a map of logging plans in public forests in Poland (https://mapy.lasyiobywatele.pl/zanim-wytna-twoj-las.html) and was looking for the best ways to incorporate time-lapse of historic aerial and satellite imagery into the map to visualize the historical context of current forest logging plans. Clearly, aerial imagery is a straightforward and intuitive visualization of executed forest loggings.

Appsilon interviewed me to establish the expectations and the budget for the launch and maintenance of the historical aerial imagery implementation. Appsilon's Machine Learning Lead proceeded to research the available data sources and integration solutions and prepared a report outlining the costs of particular solutions along with recommendations of the scenarios best suited to my requirements.

Appsilon's Senior Data Scientist & Project Leader took on the project manager role and took care of all the details from agreeing the rules of pro bono engagement, through kick-off, through final wrap up. I was very much impressed by Appsilon’s efficient, no fuss, getting-things-done attitude.

How many resources from the vendor's team worked with you, and what were their positions?

I worked with Senior Data Scientist & Project Leader to agree on the scope of the work and with Appsilon's Machine Learning Lead on the project proper.

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

Appsilon prepared a report detailing available solution scenarios with cost estimates and recommendations. Based on the report I have decided to integrate our map with aerial imagery provided by Google Earth Engine and this functionality has been since then implemented and is live at http://mapy.lasyiobywatele.pl/zanim-wytna-twoj-las.html . The report has not been released into the public domain, but we are happy to share and/or discuss the report with Clutch to confirm and demostrate Appsilon’s work product.

How effective was the workflow between your team and theirs?

Exceptionally effective. The Senior Data Scientist & Project Leader took the project manager role and was making sure that the scope is defined, expectations are managed and agreed upfront, the meetings and milestones are agreed and scheduled. The Machine Learning Lead provided estimates on the delivery and wrapped up the delivery within the promised time frame. The Senior Data Scientist & Project Leader took care of proper project closure, feedback, follow-ups. I was explicitly pinged and contacted when there was a delay on my end. I can attest to exceptionally professional and effective workflow and great project management at Appsilon.

What did you find most impressive or unique about this company?

Professionalism. Straightforwardness. Project management and getting things done. Pace & timing -- clearly my contacts were immersed in a fast-paced environment and my project was by no means the only one on their minds, yet I felt that I was getting a lot of attention, my contacts were fully present in the meetings and the meetings provided sufficient space and were not rushed.

Are there any areas for improvement or something they could have done differently?

None that I can think of.

BACKGROUND

Introduce your business and what you do there.

I’m a research data analyst at Johns Hopkins University, specifically for the School of Public Health.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Appsilon?

We were tasked with developing a website and dashboard to track tick-borne diseases like Lyme disease. Our team needed help with developing the dashboard component of that data presentation. 

SOLUTION

What was the scope of their involvement? 

Appsilon developed a static website using the Gatsby framework to house our dashboard. The dashboard itself was built in R Shiny, and we also used a combination of JavaScript typescripts. 

We wanted to make a solution that was similar to the COVID-19 dashboard that Johns Hopkins had developed. There needed to be a map embedded in the website that people could go to for reference.

What is the team composition? 

We initially worked with a salesperson to discuss the project logistics. Then, we moved into working with the president of the company and our assigned project manager who coordinated with the different coding engineers involved.

When we actually got to the two-month development phase, we had a project manager, a designer who figured out how everything would look, and two primary coders: one for the frontend and the other for the backend dashboard.

How did you come to work with Appsilon?

We pretty much knew that we wanted to build the dashboard with R Shiny, and there was a limited number of companies that specialized in that technology. We reached out to R Studio, which was a company that sold products and services related to the language, and they suggested a small list of companies, including Appsilon.

From there, we researched Appsilon and found that they were the only company that had demos that we could look at. The other candidates just didn’t showcase their work the way Appsilon did.

How much have you invested with them?

The total cost was about $60,000–$70,000. 

What is the status of this engagement?

We reached out to them back in December 2020, but their schedule was so booked! We officially did the development with them between March–May 2021. They still provide some ongoing support services for maintenance and the like.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

Our main requirement was to have the product done by our launch date of May 3, 2021, and they succeeded on that front. In terms of use metrics, we’ve gotten positive feedback from users. We’re still tracking usage through Google Analytics and have seen pretty good results.

How did Appsilon perform from a project management standpoint?

We were really happy with Appsilon. They were always in communication with us. We had weekly meetings, but there was also an online project board on GitHub that was really helpful. I could go on at any time of the day and see where they were moving specific tasks around from “in progress” to “completed.” That’s also where they’d upload the most recent version of the code.

It was a very tight project timeline, too. We were nervous throughout because we weren’t sure if we’d have enough time. However, their team was good at making us feel like they were going to make it. On top of that, we never really had any doubts about their coding capabilities.

What did you find most impressive about them?

They were just a great group of people to work with. Personality-wise, whenever we hopped on the phone, they seemed to have a lot of interest in the project because they also had issues with ticks where they were based. That level of attention showed us that they wanted to make sure our project succeeded. They didn’t give us the impression that they were going through the motions to code the site.

Generally, they were responsive to our feedback. There was a lot of back-and-forth for a while on the visual aspects of the site, but they were good about replying to us within our tight time frame. That was particularly notable since they were in a totally different time zone and country. 

Are there any areas they could improve?

This isn’t a reflection on them necessarily, but the project did feel a bit rushed because we weren’t able to start until March. It took a little while for the site design to come together because the way the site looked was really important to us. In the end, we were really happy with the design, but we didn’t have a sense of how it would turn out until a couple of weeks before launch.

Do you have any advice for potential customers?

Contact them as early as possible. Dashboards have become really popular, especially recently with the COVID-19 dashboard, so their services are more in demand. Reach out to them to make sure that you can get on their schedule, especially if you have a tight timeframe. Other than that, I don’t have any reservation recommending them to anyone.

BACKGROUND

Introduce your business and what you do there.

I’m a data scientist for a supply chain company.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Appsilon?

We needed a world-class, shiny application delivered in a fairly short amount of time.

SOLUTION

What was the scope of their involvement?

First, we scoped out the project and developed the wireframes of how it could look. Then, they iteratively worked with our team to implement and build the application. We had a lot of the code already written. Appsilon took the app and made it look really nice. The last stage included the final deliverable and deployment.

What is the team composition?

We worked with six different individuals.

How did you come to work with Appsilon?

They had a good reputation based on our research. Their team seemed to build world-class shiny applications that we’d be looking for. We went through the interview process with them and a few other companies. We liked what we heard with Appsilon, and decided to move forward.

How much have you invested with them?

We invested between $100,000–$500,000.

What is the status of this engagement?

The engagement was from January–December 2020.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

The project was successful, and our team is highly impressed. Our customers are extremely delighted by the quality of the application as well.

How did Appsilon perform from a project management standpoint?

Appsilon’s project management was excellent. We very much understood what was being worked on and understood the timeframe of future work to be done. It was clear, concise, and well-communicated. Our teams utilized GitHub as well as primarily communicated through online meetings, emails, and face-to-face correspondence. An Appsilon employee flew to our headquarters when the engagement started.

What did you find most impressive about them?

The quality of their work is extremely good. We were also impressed with the communication about what was being worked on as well as their integrity and honesty.

Are there any areas they could improve?

No, they did everything for us plus more than what we expected.

Do you have any advice for potential customers?

Take a look at their online portfolio to get an idea of the type of work they do so you have an idea of what you might like when you talk to them.

BACKGROUND

Introduce your business and what you do there.

We’re a small software startup, and I’m the chief of staff. I help with the product roadmap, organizational development, and more.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Appsilon?

They helped with RStudio development, which is an open-source software development environment.

SOLUTION

What was the scope of their involvement?

In the first phase, they developed a web application using Shiny, which is well known in RStudio development tool. There have been multiple tasks, but the main one is to improve the UI of one of the tools on our platform.

The tool is web-based, and Appsilon helped us deliver a text message scheduling feature. They also redesigned the application, built a new log-in page, and created components that we could use in the future to train users who are less experienced with Shiny development.

Initially, we provided a project plan, as well as some charts, pictures, and ideas around UX. We pointed them in the right direction, and Kamil (Full Stack Engineer) contributed some ideas. It was a joint effort.

We’re starting up again next week with the second part of the project.

What is the team composition?

We’ve worked with two developers, one being Kamil, and Anton (Head of Sales) is our salesperson. Anton handles all the operational aspects of the project.

How did you come to work with Appsilon?

We decided to partner with them because of their experience with Shiny. We specifically wanted people well-versed in building in that environment, and they’re one of the most well-known Shiny contractors.

They were already selected when I came on board, and our CEO knew of them previously.

How much have you invested with them?

We spent around $60,000–$70,000 on the first phase.

What is the status of this engagement?

The first phase ran from December 2020–until March 2021, and phase two kicks off in April 2021.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

We haven’t put the application into production yet, but we’re really happy with the outcome and how it looks. Kamil was strong at recommending behind-the-scenes components to build that will help us in the future, even though they weren’t critical at the time.

How did Appsilon perform from a project management standpoint?

The lead developer was really good at project management. We worked with him directly, and we could run with a lot of tasks without having to involve the CTO, which was good.

Initially, I gave a project plan, and then they developed a more in-depth version. We aligned our two views and moved forward. The second time, we used the same approach but implemented a GitHub board to agree upon priorities.

What did you find most impressive about them?

There is a strong trust element between us. By the end of phase one, I really trusted Appsilon with my ideas. I brought ideas to them and they came back with recommendations that I felt placed our best interests first. Kamil is definitely a trusted partner.

We’re switching up our people a little bit for this second stage, because of their availability. Kamil is moving from a full-time posting with us to a part-time posting — we didn’t want to lose him.

They are based in Poland, and we’re based in California. My team works pretty late, and they’re flexible, but organization is critical when face-to-face time is infrequent. Organization helped us be successful. We wouldn’t have been successful working with just any group because of the time difference.

Are there any areas they could improve?

They really improved over time in terms of project planning. Looking forward, it would have been nice to be able to meet the person we’re going to be working with before signing the contract for phase two.

Do you have any advice for potential customers?

In-depth project planning will be very helpful to you. It was the easiest way for us to showcase our goal.

BACKGROUND

Introduce your business and what you do there.

I’m a director at a big pharma company.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Appsilon?

We needed software development services.

SOLUTION

What was the scope of their involvement?

Appsilon did a full frontend software development lifecycle in agile sprints. They did the coding using R Shiny, and they also did the testing and deployment. It’s a data analytics solution and frontend interface to reduce the effort and cost in preparing key documents.

Additionally, they’re helping us install CI/CD through the use of the Jenkins package, which will allow us to do automated testing.

What is the team composition?

We work with a team of three developers.

How did you come to work with Appsilon?

Someone else on my team found them, and we chose them because of their expertise with R Shiny apps.

How much have you invested with them?

We spend about $1.5 million annually. 

What is the status of this engagement?

Our partnership with them began about three years ago, in 2017, and it’s ongoing. 

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement? 

We have rollouts of new versions of the software. We look at the number of improvements and their ability to do the work in the appropriate timeline, and they’ve done a good job of getting things done on time. 

They are nimble when it comes to updating the code, providing new features, and identifying errors and bugs. We have weekly and daily meetings.

How did Appsilon perform from a project management standpoint?

We manage the project, but they use all the expected tools, including Jira, Confluence, and BitBucket — the whole Atlassian suite. 

They are very flexible about altering the work from one sprint to the next, which is the point of the agile methodology. Appsilon has shown a lot of initiative in terms of finding problems and fixing them, which is helpful. We communicate using Microsoft teams.

What did you find most impressive about them?

Their speed and flexibility are impressive. For example, when we have an issue on a particular day, by the time we come back the following day, the issue is resolved because of the time difference. 

Are there any areas they could improve?

Initially, they missed some bugs, but Appsilon recommended that we adopt automated testing with Jenkins, which will help solve that problem.

So, there was an issue, they suggested a better solution, and now we’re implementing it. This will help all of the applications we’re working with. They did a great job of identifying and fixing the issue.

Do you have any advice for potential customers?

They're great service providers and partners. I highly recommend Appsilon.

BACKGROUND

Introduce your business and what you do there.

I’m the VP of data and analytics at SeQuel Response. Our company is an all-encompassing direct response and advertising agency.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Appsilon Data Science?

We had an online tool that we were looking to rebuild and reconfigure. We wanted to use our studio because a lot of our data scientists are familiar with Python. So, we brought on Appsilon Data Science to help us develop a real-time online analytics tool using our own Python.

SOLUTION

What was the scope of their involvement?

We have a direct mail campaign. Appsilon Data Science built a tool that takes the customer file and matches it back to the mail file that we mailed to determine how many we actually mailed and could take credit for mailing in that situation. 

From there, it actually rolls up the response rate or customer sales rate on the list we mailed and the different creative packages, and it gives an all-encompassing dashboard. 

What is the team composition?

We worked with between 6–10 different teammates, but we mainly interacted with 3–4 main people.

How did you come to work with Appsilon Data Science?

We reached out to our studio and got a list of preferred vendors. We met with several of them, and we chose Appsilon Data Science because they fit our needs the best and their pricing was what we wanted. 

After meeting with them, it was obvious that they were the team we needed to work with because they understood our work and asked good questions.

How much have you invested with them?

We spent a little over $30,000. We also have a studio license from them that was a little over $18,000. 

What is the status of this engagement?

We signed the initial agreement with them in April 2020. Although we ended this project in September 2020, we’re still going to need some help with the development. 

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement? 

The quality of the work and service was as high as it could’ve been. 

How did Appsilon Data Science perform from a project management standpoint?

We corresponded mostly via email because of the time difference, and we also had weekly online and in-person meetings. Appsilon Data Science was outstanding in terms of deadlines. 

We made a lot of changes on a regular basis, but the team was great at staying on-track and getting stuff done within the timeline they thought was reasonable. They were also very open and honest, so if they thought something couldn’t be done within a certain timeframe, they let us know.

What did you find most impressive about them?

Their knowledge was outstanding. A lot of this was unknown to us, but Appsilon Data Science was able to do their work efficiently and quickly. Their overall demeanor and personality were great as well.

The team always seemed happy and willing to work. In those two areas, they were above and beyond anyone we could’ve worked with.

Are there any areas they could improve?

No, I have no suggestions. Everything ran smoothly with them. I’m happy with the end result. They were a great team to work with and they should just keep doing what they’re doing. 

Do you have any advice for potential customers?

Be open and honest with Appsilon Data Science about all facets of the project. They can’t help you unless they know what you’re thinking. 

BACKGROUND

Please describe your company and your position there.

I am the Product Developer for a Fintech company based in Dublin (Ireland).

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire Appsilon Data Science?

We had to bring our reporting capabilities to another level. So far everything was based on a manual process, which mainly rely on multiple excel files. The entire process was long, highly fragmented and inaccurate, so we needed to outsource the entire redesign and rebuild.

What were your goals for this project?

We wanted to have an automated reporting process, which give us the full controls of data and final reports. We had to replace manual and disconnected reporting workflow with fully automated ones.

SOLUTION

How did you select Appsilon Data Science?

We come across Appsilon Data Science during an online conference (eRum 2020). Appsilon clearly was the one who could offer us the best services.

Describe the project in detail.

We worked remotely, we never met in person. Nonetheless, everything worked fine. Appsilon has been since the beginning, well organized and highly collaborative. They delivered on time, perfectly managing different consultants all along during the delivery process. Highly recommend.

What was the team composition?

Project manager, data engineer, data scientist, data visualization expert.

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

Initially, we were assigned a project manager to design the final apps and serve us us the main point of contact during the whole project. More than 3 consultants have been involved in the final delivery, leaving us really happy about the choice made.

How effective was the workflow between your team and theirs?

In one word: perfect!

What did you find most impressive about this company?

Young, well prepared, incredible well organized, friendly approach, open, honest, innovative.

Are there any areas for improvement?

I am sure there are ... but we were not able to see any.

BACKGROUND

Please describe your company and your position there.

I am the Principal Modeller and Programmer at a health economics consultancy based in the UK.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire Appsilon Data Science, and what were your goals?

We had developed a R Shiny application in house, however we needed assistance with some software engineering tasks. We required an authentication layer to be added to our application allowing us to control access, alongside an easy way to deploy.

SOLUTION

How did you select this vendor and what were the deciding factors?

We came across an article online which discussed a deployment solution Appsilon had developed for another client. The problem sounded like ours, so we decided to get in contact.

Describe the scope of work in detail, including the project steps, key deliverables, and technologies used.

There were two iterations to our collaboration. In the first phase, Appsilon Data Science constructed a workflow allowing us to easily deploy our application, built a staging environment to facilitate quality assurance, and added an authentication layer so that we could control access.

The solution used a combination of technologies: Ansible, ShinyProxy and Auth0. We were very impressed with the quality of work and rigorous approach, so we then hired Appsilon Data Science for a second phase where they refactored existing code to significantly speed up performance of in-app calculations.

How many people from the vendor's team worked with you, and what were their positions?

We predominantly worked with 2 people from Appsilon, a data scientist who also managed the project, and a software engineer. We were also in communication with an account manager.

RESULTS & FEEDBACK

Can you share any measurable outcomes of the project or general feedback about the deliverables?

All objectives of the collaboration were met within the timeframe agreed. Appsilon Data Science also added a lot of extra value by providing additional recommendations on code structure and ways of improving the performance of the application.

Describe their project management style, including communication tools and timeliness.

Appsilon Data Science had a very methodical and rigorous approach. The project began with a kick-off meeting where they raised questions about the deliverables. A way forward and timelines were then agreed.

They were very communicative throughout the process both via email and video calls where they demonstrated the delivered solution. They also provided comprehensive documentation and were very responsive to any queries raised.

What did you find most impressive or unique about this company?

The high quality of their work and excellent communication made them a pleasure to work with. We hope to embark on further collaborations in the future.

Are there any areas for improvement or something they could have done differently?

No

BACKGROUND

Please describe your company and your position there.

We are a company that specializes in working with professional athlete data. I am the director of analytics for our company, in charge delivering projects related to product + analytics.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire Appsilon Data Science, and what were your goals?

We hired Appsilon to build the prototype for our new web application. We already had it designed, we needed help getting the project off the ground so that it could then be handed off to our internal team. We needed an MVP quickly, in short.

SOLUTION

How did you select this vendor and what were the deciding factors?

We met Appsilon at an industry conference and our #1 deciding factor was their expertise in the tool/language we required - RShiny

Describe the scope of work in detail, including the project steps, key deliverables, and technologies used.

The main technologies used were RShiny and GitHub. We delivered the design and required features to Appsilon for the app, and they were charged with delivering a fully deployed app that met all of these requirements.

How many people from the vendor's team worked with you, and what were their positions?

Over 5 people worked with us across all levels of technical experience (junior to VP), along with their account manager. All fantastic folks to work with !

RESULTS & FEEDBACK

Can you share any measurable outcomes of the project or general feedback about the deliverables?

It was delivered on-time, under budget, and met all our requirements.

Describe their project management style, including communication tools and timeliness.

We had sprint meetings 3x a week in addition to a slack channel. Communication was always 100% clear.

What did you find most impressive or unique about this company?

The technical prowess of their engineers is exemplary. They are able to write elegant, clear code that makes difficult requests seem pedestrian.

Are there any areas for improvement or something they could have done differently?

No.