Unoperate

Introduce your business and what you do there.

I’m the CEO of TEONITE Venture Builder, a software development company.

What challenge were you trying to address with Unoperate?

We needed someone to fill the gaps in our software development efforts.

What was the scope of their involvement?

We hired Unoperate for their low-level coding and device integration expertise to supplement our communication devices in the telecommunications space.

What is the team composition?

We worked with only one expert from their team.

How did you come to work with Unoperate?

I’ve personally known Marek (Founder) for quite some time now, so I’m aware of his technical capabilities and knowledge.

How much have you invested with them?

We invested about $10,000.

What is the status of this engagement?

We worked together from June 2019–March 2020.

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

The partnership was relatively easy because they successfully delivered the specific requirements we wanted.

How did Unoperate perform from a project management standpoint?

All their tasks were delivered within the given timeline and they consistently communicated with us throughout the duration of the partnership. Unoperate was highly persistent in pursuing our time frames, which was really important for me.

We utilized JIRA to manage the project.

What did you find most impressive about them?

We were impressed with their technical capabilities, communication skills, and deep knowledge and expertise.

Are there any areas they could improve?

I don’t have any constructive feedback on how the project was handled.

Do you have any advice for potential customers?

To get the most out of your engagement with Unoperate, I advise you to involve Marek in the entire process.

Introduce your business and what you do there.

I’m an engineering manager in Google’s cloud division. My team operates Cloud Bigtable, which is Google’s petabyte-scale NoSQL system. 

What challenge were you trying to address with Unoperate? 

We wanted to make HBase migrations easier for customers. For context, HBase was inspired by the Bigtable paper that Google published in the early 2000s. It was open-source, and plenty of companies wanted to move into the cloud. As a result, we had a lot of market penetration, and Bigtable was the obvious choice for customers running HBase clusters. We wanted to make their HBase migration to Bigtable easier and avoid taking any downtime. As a result, we hired Unoperate. 

What was the scope of their involvement?

Our team engaged Unoperate to create an open-source product for customers to use for their migration from HBase to Bigtable. We previously connected to the database using a specific piece of software. Unoperate essentially created a new version of that software that could talk to both HBase and Bigtable without requiring end users to take any downtime. 

They had to solve plenty of challenges, specifically in terms of what would happen if the program failed in one of the databases. HBase and Bigtable were completely separate systems; there was no consistency between them, and the software might succeed in one of them but fail in the other. 

This posed an interesting challenge for Unoperate, which had to consider correctness, concurrency, and compatibility across different database versions. Overall, our team gave Unoperate quite a few challenges, and they worked to solve them all by creating a piece of software that was entirely Java-based. 

What is the team composition?

We worked with their lead architect and lead developer. I mainly interacted with these two members, but there might have been other people involved in the project. 

How did you come to work with Unoperate?

Unoperate was run by a former Google employee, who was well-respected within the team. That person also used to work with a sister team, and people knew his outputs well — he was a high-quality engineer. As a result, we decided to hire their team for this project. 

What is the status of this engagement?

The project lasted from June 2021–March 2022. 

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

A couple of engineers from our team worked closely with Unoperate. While their resources did their own code reviews, our team also reviewed what they wrote. We spent some time going through multiple rounds of evaluation and testing. This process gave us confidence in what they’d built — we knew we could put it in front of customers. Moreover, the feedback for Unoperate’s work from our engineers was great. The testing and customer feedback were positive. 

How did Unoperate perform from a project management standpoint?

Unoperate managed the project well. We overshot the initial budget estimate, but we didn’t put the blame on them. We simply faced challenges that we didn’t anticipate. 

Meanwhile, we essentially let the team manage the project on their own. We had a shared Google Document, and our teams shared the code on GitHub, which was our preferred means of collaboration. We also didn’t want to manage any specific subtask they had. Moreover, they were using a project management tool internally, but we didn’t get involved in that. 

What did you find most impressive about them?

For this project, we took quotes from a few other teams, and Unoperate’s price was in the middle. However, they were much more thoughtful than other teams about the project and what it would take to produce the software. 

On top of that, Unoperate put plenty of time into testing, documentation, and sample code. They accounted for dealing with different HBase versions. Overall, the level of maturity and experience they demonstrated in their code was clearly better than others. 

Are there any areas they could improve?

If we had to pick one thing that didn’t go according to plan, it was the delay in the timeline — the project took longer than we hoped. We wanted to wrap everything up by December 2021, but we only recently finished. 

To be fair, much of the delays were on our end. Our members who were doing the code review were busy and unavailable. Other than the delay, there were no other areas of improvement. Anyone working in software development projects would know that estimation could be extremely difficult. In the grand scheme of things, going over by a few weeks wouldn’t be out of the ordinary. 

Do you have any advice for potential customers?

The more you can clarify and write down upfront, the fewer surprises there will be when you start the engagement. At Google, we follow a process of writing SOWs, which go into a fair amount of details on what the deliverables are, what we’re expecting in terms of code quality, test coverage, and similar things.