Mutt Data

Please describe your company and your position there.

’m the Software Development Manager for the largest online commerce and payments ecosystem in Latin America. We enable e-commerce and digital and mobile payments on behalf of our customers by delivering a suite of technology solutions across the complete value chain of commerce. We are present in 18 countries.

For what projects/services did your company hire Mutt Data?

The company needed to improve the current sponsored listing system, conciliating ads revenue maximization with seamless marketplace (i.e. organic) listing integration. The key was for the solution to avoid being detrimental to the marketplaces functioning or traffic.

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

Through a referral, they had a proven track record working with trusted companies/colleagues We met with different vendors. Their engineers were great communicators /very knowledgeable. Most importantly, Mutt Data has a great culture and working with them is very easy. Mutt Data fit the skill and budget profile we needed for our project.

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

The project was organised into clear phases. During the Discovery Phase, the vendor went over our current status, data, infrastructure and business goals. With a milestone based plan and roadmap we moved on to KickOff where all the necessary tools and communication channels were set up. After that, the vendor dove into development, implementation, deployment and hardening and finally what they called Knowledge transfer.

The solution consisted of creating a platform over which our company could rank advertisers, decide on the cost of advertising spaces and filter bad quality ads in order to sustain organic GMV on the platform. During the project the vendor proposed and implemented a Conversion Rate Estimator and Click-Through Rate Estimators.

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

We worked with about 6 dedicated resources. The members had different positions and seniorities but there was an assigned Tech Lead dedicated specifically to the project and we had frequent contact and follow-ups with two of the companies founders (CRO and CTO).

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

Costs were reduced significantly while maintaining target ROI and GMV. The developed system is currently functioning with improved results. All deadlines were met.

How effective was the workflow between your team and theirs?

Their team was very communicative, follow-ups were frequent and there were clear processes and tools in place to maintain an efficient workflow. The team was always available, and we had no hiccups.

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

• Communication Skills & Strong Culture based on continuous learning Their team had been carefully selected for the project at hand, they were very dedicated, great communicators. The solution was clearly tailormade.
• Their team was great to work with, they really took the time to understand our current situation in order to work on a lasting solution.

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

• No clear areas for improvement stand out.
• The vendor is open to suggestions and changes.

Please describe your company and your position there.

I'm an AdTech Engineering Manager of an international technology company, which creates leading multi-platform brands that entertain, connect and add business value around the world. Our Gaming division has exceeded 800 million downloads and our Brand Gamification division is a pioneer in the region, offering marketing and advertising solutions in the mobile game universe. The company also applies AI to enhance digital transformation.

For what projects/services did your company hire Mutt Data?

As one of the largest trivia game publishers in the world, etermax has a significant mobile ad inventory to monetize. One of the ways in which etermax's games are monetized is through in-game ads that are sold to several 3rd party Ad Networks through the available mechanisms: Mediation and Real Time Bidding. Mutt Data was hired to help us design and implement optimization mechanisms around our Mediation Platform.

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

They were recommended to us by a trusted 3rd party. Their ideas were innovative and they were able to effectively communicate their expertise in the subject. Their track record working on similar companies was also a deciding factor for us.

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

The project was separated into a series of phases each one with its purpose, milestones and results. We first started by a research phase which allowed the vendor to get an understanding of the problem that we were trying to solve and then designed, together with our team, a first solution to start implementing. The final solution consisted of a Machine Learning based strategy which allowed us to tailor, in real time, an Ad Waterfall for each user in order to maximize revenue. As a complement of the solution, the team also created an MLOps platform setup to enable rapid iteration, tracking, deployment and testing of Machine Leraning model variants.

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

We worked directly with 4 members of Mutt’s team, most of which were Machine Learning Engineers. One of the company's founders also participated in overall design and overview of the project.

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

While the optimizations are still being iterated and tested in different games and audience segments, in some cases it has increased revenue by approximately 5% over already manually optimized waterfall configurations.

How effective was the workflow between your team and theirs?

The teams were able to work together and communicate without any issues. Communication channels and workflow processes were clear from day one. Some of the tools used include: Slack, ClickUp, GSuite, Gitlab, among others.

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

The team is always available and they take the time to think solutions out taking specific client constraints and goals into account. I appreciated that the solution was tailor-made to our needs. It didn’t feel like it was just another project or that we were outsourcing some software module but rather that Mutt's team was an extension of our internal team, and worked alongside them in a very fluent and natural way.

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

Nothing particular comes to mind.

Introduce your business and what you do there.

I’m the CTO of First Delivery, a delivery logistics software company for third-party ordering services. I manage our technical team and day-to-day software operations.

What challenge were you trying to address with Mutt Data?

We needed help forecasting demand for restaurants in several cities around the country.

What was the scope of their involvement?

Mutt Data built a machine-learning model that helped predict demand for restaurant delivery orders. The model also predicted the necessary staffing capacity so we’d know how many drivers we’d need to onboard. 

During the discovery phase, we provided the project scope and shared our data structure and business problems for Mutt Data’s understanding. We did some exploratory research together and they uncovered insights we hadn’t been previously aware of. In the next phase, we onboarded them onto our technology stack and set them up to have their software interact with our stack. They used a set of open-source machine-learning and data engineering resources, including Airflow and MLflow. We also provided them with some Amazon resources.

We hosted the model but they helped maintain it by upgrading it and fixing bugs. 

What is the team composition?

I started off the project with Juan Mateo (Co-Founder & CRO) and also worked with Alejandro (Knowledge Lead), Gian Franco (Data Developer), and another teammate.

How did you come to work with Mutt Data?

I initially found them through a search engine run by a business friend of mine, who recommended them when I asked about them. 

We selected Mutt Data because they had the capacity to start the project within a reasonable amount of time and their pricing was competitive. They had worked on similar projects and it seemed like a good match. I also got along well with Juan Mateo. 

How much have you invested with them?

We spent approximately $200,000 in total over the course of the relationship. The initial project cost was $40,000–$50,000 but we continued to work with them.

What is the status of this engagement?

The project lasted from March 2019–March 2020.

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

We reduced our weekly operations spending by 25%–35%. 

How did Mutt Data perform from a project management standpoint?

Everything was delivered on time, and they made sure to communicate everything clearly. They tracked daily billable hours in Excel. They were really helpful and quick to respond when small issues arose that we couldn’t fix; they were always online to solve those on short notice. The majority of the communication was in a shared Slack channel, which we actually still have to this day. Weekly status updates were communicated over the phone. 

What did you find most impressive about them?

Mutt Data understood the problem we were trying to solve from a data science perspective and delivered on promises. They addressed a high-level business problem and implemented software solutions that are still delivering to this day. I’d recommend them to anyone with data science needs.

Are there any areas they could improve?

No, they were awesome.

Do you have any advice for potential customers?

Make sure that whoever’s working closest with Mutt Data has a solid understanding of the business domain. When they start to build the project out, that point of contact should be able to manage the project scope and communicate what is and isn’t needed.

Introduce your business and what you do there.

I’m a senior manager at an e-commerce company. 

What challenge were you trying to address with Mutt Data?

We built a programmed shopping platform and needed Mutt Data to support the process of building it. Apart from that, they collaborated with us on developing an audience management system. 

What was the scope of their involvement?

We’ve done two software development projects with Mutt Data. Our teams have created documentation together, and they’ve been involved in the project since its initial stages. The shopping platform has an automated system that handles large volumes of data. There’s machine learning (ML) technology embedded in it, so it’s a complex system. The main coding language we’ve used is Python. 

What is the team composition?

We’ve worked with 10–15 people, and the team composition has changed throughout the year. Their members include developers, backend experts, project leaders, ML engineers, and more. 

How did you come to work with Mutt Data?

Mutt Data was already hired by our company when I joined it. However, I knew the team from the past; I worked with them at another company, so I considered them to be a great choice of vendor. 

How much have you invested with them?

We’ve spent around $1 million. 

What is the status of this engagement?

The project started in December 2021, and it’s ongoing.

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

Our most important success metric is that we can go into production within our expected timeline. We’ll be able to assess the performance metrics once the products are launched. Nonetheless, this has been a long process, and Mutt Data has helped us a lot in terms of generating ideas — they have great knowledge. 

How did Mutt Data perform from a project management standpoint?

The management of the project mainly comes from our side, but Mutt Data has been in charge of project execution. We use Jira for project management and Slack, Google Meet, and Zoom for communication. 

What did you find most impressive about them?

Their greatest strength for this project is their knowledge of advertising. Not many development companies have such good insight into that area, making them stand out from their competitors. 

Are there any areas they could improve?

There’s a lot of room for improvement in terms of the project’s execution. 

Do you have any advice for potential customers?

If you’re using the same scheme as we do, try to integrate Mutt Data with your team as soon as possible. Make that integration as smooth as possible. 

Please describe your company and your position there.

I’m the Founder of a Latam based, Information Technology Solutions Consultancy company that combines innovation, agile design methods, and technological development to deliver high-quality and tailor-made digital products & integrations. We specialise in front-end, back-end, ecommerce platforms, integration services and devops and databases.

For what projects/services did your company hire Mutt Data?

We needed to streamline our data processes, updating our data platform and components to improve efficiencies in storage, transformation and scalability. This required implementing new tools, processes, workflows and paying off technical debt.

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

We selected the vendor based on trusted reviews from industry colleagues who highlighted their expertise in Machine Learning and Big Data. We knew one of the company’s founders well, and got in touch to brainstorm ideas for our challenges. We were convinced by their suggestions and knowledge on the matter. The selected vendor had a great track record, had worked on many projects and was in-line with our expectations in terms of quality and budget.

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

The project had predetermined stages and they were all complete in time and form. There were six main stages. Each had its clear set of goals, milestones and review dates. The whole process was very organised and clear.

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

We worked with a team of three resources full time. One of them was the lead and our main point of contact. One of the company’s founders was also involved in project reviews and status. The team was qualified, efficient, communicative and easy to work with.

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

We saw a 35% reduction in data delivery times inside our organisation. The implemented system reduced data processing and delivery times drastically. The freed up time can now be allocated to current and potential client projects which generate value for our company.

How effective was the workflow between your team and theirs?

Easy to work with. Good Workflow. We were also aligned on the tools and framework used to structure the project’s process: agile, scrum, asana, notion, slack, etc. We got along very well with the vendor team, they were easy to communicate and work with. We had no issues working with the vendors team. They are experts in this field, and looked to tailor their recommendation based on strategic advice and, later, with effective implementation

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

We were positively impacted by their team's startup working culture. They genuinely seemed to enjoy the challenge. They were also very agile. A lot of companies mention AI, ML, and Big Data as part of their services but the vendor truly showcased unique expertise in these fields. They are very up to date with the latest practices and tools, and expertly trained.

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

Satisfied with the project. No comments.

Please describe your company and your position there.

I’m the Chief Technology Officer (CTO) for an consumer education technology company. We operate globally in more than 180 countries with a community of more than 50 million teachers and families who come together to share kids’ most important learning moments in school and at home– through photos, videos, messages & more.

For what projects/services did your company hire Mutt Data?

We needed a Self-serve DataOps Platform that would empower engineers and analysts to make data processing, storage, transformation and availability, scalable and efficient. We needed to modernize our data platform, addressing ETL issues, duplicated extraction processes, low-confidence in the data, troubled data ingestion from multiple sources and a improve testability and monitoring of input and output tables.

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

Internal referral made in Y-combinator's internal forum, from startups founders that had previously worked with them.

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

The project began with a discovery phase where the vendor took the time to understand our company, status, data, infrastructure and business goals. This was all summarized into a visual plan with proposed solutions and a clear milestone-based roadmap.

After that, the project was organized into the following phases: Kickoff (tools, system access and project organization/communication setup), Development, Implementation, Deployment and Hardening and finally Knowledge Transfer (making sure our team could use and own the new platform).

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

Their CRO was our main pain of contact, but we worked together with a multidisciplinary team of 3 experts. The team had a dedicated Tech Lead and team specifically assigned to our project.

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

By the end of the project there was a 10x increase in data pipeline processing and data asset deliveries. The end result was a data platform that can handle all data processing, transformation, lineage, governance, self-serving workflows, and AI enabled alerting and monitoring for our company’s terabytes of data.

How effective was the workflow between your team and theirs?

Communication was fluid and frequent, we mostly communicated over video chat but also made heavy use of selected tools such as: Asana, Slack, Google Workspace, and Github. The team was open to questions, and quick to answer. Milestones and timeframes were accomplished.

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

- Agility to find new ways to solve problems as they developed. - Their experience is noticeable. They had a lot of ideas that were sparked by different projects and industries they had worked in. - I appreciated that it didn’t feel like this was just another project, the team was truly involved in every single aspect of the project. A very dedicated team.

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

They could have been more proactive in promoting adoption of the new tools they were building across the organization. It took some effort to get that new knowledge out to the rest of the teams. We also got to the end of the initial roadmap, but didn't have a plan for a phase 2 ready.

Please describe your company and your position there.

I’m the CEO for a startup dedicated to software development with a focus on financial audit automation. Currently, our company offers services in business intelligence, data administration, machine learning and process automation for financial departments. We have a strong presence in the banking industry.

For what projects/services did your company hire Mutt Data?

We were looking to implement a complete fraud detection solution for our bank auditing efforts. This included setting a system in place to identify anomalies in regards to loss control, internal scams, false positive fraud cases, and investment fund allocation control. The goal was for the system to reduce false positive fraud and reduce fraud generally speaking in our biggest clients.

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

We heard about the vendor on social media, local media outlets and from trusted colleagues who recommended the company to us. The recommendations were a deciding factor. The vendor was in our short list, the deciding factor was the showcased expertise in the discovery meetings. The vendor ticked all of our boxes, they came recommended, they had the necessary team and ml knowhow and they were within our estimated budget.

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

The project was organized into different stages. At first we had a series of review meetings where we analyzed our painpoints and possible solutions to them. Then this was translated into a plan. After that, the vendor set up some channels and tools to start working. The next stages were basically the development of the solution itself, more reviews, improvement stages and final delivery.

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

There were 5 resources from the vendor’s team. There was a senior data engineer in charge of the team, another senior data engineer, 2 Semi Senior data scientists and one founder who participated as well.

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

The implemented system allowed us to reduce false positive fraud cases by more than 10%. Fraud in general was lowered by around 5% in our biggest client segment. We also saw improvements in efficiency of processes.

How effective was the workflow between your team and theirs?

They’re easy to work with, responsive and proactive. They always had useful feedback and were very communicative. Our collaboration was without hiccups. There was always a clear set of stages and milestones, and a visible review calendar to keep work running smoothly. Clear work processes. Good communication and use of tools such as G Suite, Slack, ClickUp, Github, etc. *What did you find most impressive or unique about this company? They took the time to understand what we needed so they could give us the best suggestions before we decided on a proposed solution. Working with them was very easy, they felt like a part of our team.

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

We were very content with the quality of Mutt Data’s work. We’ve been using the implemented system for an extended period of time now and were even able to scale it. We noticed immediate improvement in key metrics when implemented. When the project finished, they offered suggestions on how to continue improving the system and how else we could use machine learning in the future. All in all we were very satisfied with the solution, its implementation and the team.

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

I was pleasantly surprised in this regard. Mutt Data made sure we were always in the loop and scheduled different control instances so we could review progress, changes and goals as the project advanced

Please describe your company and your position there.

I’m the Co-Founder and CTO of the largest redistributor of surplus medicine in the USA. Powered by technology, our company helps organisations like nursing homes, pharmacies and manufacturers to donate their unused medicine and get it to where it’s needed most.

For what projects/services did your company hire Mutt Data?

In order to truly leverage our data for decision making we needed to develop our first Data Warehouse where we could consolidate data from all our sources in one place. This was a necessary step in order to become a data-driven company. We needed an experienced vendor to advise us on the best possible solution considering all tradeoffs.

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

Their team came highly recommended. We felt comfortable working with a company with their positive track record.

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

The vendor began by taking the time to understand our current context, this included talking about our infrastructures, current pain points, needs, and data sources, teams, and processes. This was all part of what they called the Discovery Phase. After that, we moved on to the setup of different technical and communication tools. Finally development itself began, followed by the implementation and deployment of the solution. In regards to the solution itself: the implementation consisted of implementing Airbyte to consume data from MySQL, Quickbook, and other sources. DBT for transformations, Airflow for orchestration and Quicksight for dashboard capabilities.

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

We generally worked together with a team of six experts who were specifically assigned to our project. Our main contacts were the team's dedicated tech lead and one of the vendor's co-founders.

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

Integrated dashboards

How effective was the workflow between your team and theirs?

From day one, working with the vendor was easy going. They made a visible effort to keep in sync and all relevant people were always in the loop. The use of weekly meetings, task tracking systems and a clear roadmap made the workflow run smoothly. Their team was always readily available.

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

I appreciated their extensive experience building cloud and data infrastructures as well as full stack machine learning systems. Not only did this allow for a robust tailored solution but also for a cost-effective one.

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

We’ve been working together with the Mutt Data team for several years, suggestions are usually applied so there are not really improvements to be made at the moment.