Welcome to Part 2 of an ongoing series with ShuttleCloud engineers discussing projects, bottlenecks, and whatever else is on their mind.

Below is an interview with Ignacio Carretero, Client Services Engineer for ShuttleCloud who lives in Madrid, Spain.

At a glance

Name | Ignacio “Nacho” Carretero
Role | Client Services Engineer
Background | Manufacturing, Process mgmt, Green energy, GE – OMLP

Introduce Yourself

I studied Industrial Engineering with a specialization in Electrical Engineering. My final project in university was to build a spreadsheet model that predicts the viability of installing an electrical substation. This is a constant pain point for contractors who otherwise waste weeks forming bids that aren’t executed.

After school I wanted to become even more technical, so I got involved with the Global Supply Chain Management (GSCM) organization at GE and worked in wind turbine factories. This was completely different from my training in Electrical, but I enjoyed learning about hardware and was invited to the leadership program. During a 4-part rotation I traveled Europe and the US, took six sigma courses, and learned about management, finance, and Lean Manufacturing.

After the rotations I decided I wasn’t happy in the corporate world, so I quit GE and started an advanced degree in web design and community management. I began working with ShuttleCloud after graduating from that program.

What’s your role at ShuttleCloud?

I’m responsible for simplifying and automating internal processes — this saves time for our software developers and provides customers a better experience. To do this I serve as the interface between front-end applications and the development team. The scope of my work ranges from our self-service tool to Enterprise portals for Comcast and TWC.

What are some of the biggest challenges?

My first inclination as a new hire was to implement a process management strategy similar to General Electric, but I quickly realized it was far too complex. Startups have a much more agile development environment, so documenting every process wasn’t as useful because implementations are always changing. What I was able to do was extract the idea of high level process management, perform light documentation, and focus on making changes in areas with the biggest impact.

After coming from a corporate climate that required escalation and approvals for trivial tweaks, it was uplifting and refreshing to observe how flexible ShuttleCloud is at iterating products. Our style is to own the processes we implement, and if something works better we can try it without the red tape.

What I’ve been working on lately is removing as many complex second-line support tasks as possible, ranging from error code interpretation to rerunning migrations. By pairing issues with automatic processes, we effectively ‘demote’ new issues to first-line. Early on we experienced bottlenecks in passing on difficult issues to busy engineers, but now we have a better knowledge base that lets us isolate most errors within our department. Essentially, first-line support can now do a lot of things that only second-line or engineers could do before I joined.

While we aren’t implementing six sigma per se, the Lean philosophy has been extremely fruitful. Although I came from an environment with repeatable processes that were studied and documented in detail, at ShuttleCloud we question our processes constantly.

Which technologies do we leverage for customer support?

Most support is conducted online, and for this we use ZenDesk Enterprise. In general, helpdesk software makes it incredibly easy to scale customer communication. By building specific gateways for each major client or product, we’ve significantly decreased the lag time between customer inquiries and the corresponding solutions.

Our migration volume has grown tremendously since I joined earlier this year, so naturally the demand for customer support has followed suit. By implementing tools efficiently and automating the most common internal errors, however, we’ve been able to comfortably scale our support vigilance to match.

For example, I started in March with ~30 inquiries daily from users, yet now receive only 2-3 even though our migration volume has increased 100x in the past 8 months. To manage customer data we use CouchDB and a proprietary error database repo that contains highly specific solutions for every possible migration use case.

What makes ShuttleCloud Support so great?

When migration volumes tripled in May we added another member to our support team, so we now span 2 time zones (EST / UTC + 1). By focusing on customers first we solve the most common issues overnight. If a process for a particular solution stops working for a few customers, we change it immediately to better fit the use case. From software development to customer communication, we remain very agile.

Additionally, we have open communication between Support and Engineering with no bureaucracy in between. If issues arise with an endpoint, our engineers know immediately. In this way I get to be the “eyes” of our developers, debugging problems live while engineers build them.

Describe a project you’re proud of.

Our error code repo is an exceptional project that we’re iterating constantly. Taking into account my Industrial Engineering background, I found that unifying our language for sharing customer issues has been extremely helpful.

It is by interfacing with developers and speaking clearly about problems with unified language that we’ve been able to make support issues specific and actionable. I’m very proud of this initiative.

Leave A Comment

Your email address will not be published. Required fields are marked *