Charles Wilkinson

I started my career as a Java developer at Portaltech Reply working on e-commerce projects built on the Hybris platform and maintain those skills to this day. I grew quickly to technical lead level and later solutions architect. After focusing for a number of years on the technical architecture side of the equation, I decided to broaden my horizons and so took ownership of the services (operations and support) side of the business. Taking on this new challenge on top of my existing role as a principal architect was challenging but ultimately very rewarding.

Architect

I specialise in e-commerce, with a further specialism in Hybris having architected solutions in the retail, telecommunications, travel and Fast Moving Consumer Goods verticals (both B2C and B2B). While I have not yet directly architected a solution in the grocery space, I have consulted on several grocery Hybris implementations and am very familiar with the many challenges that exist there.

I also have significant experience in the telco space. Indeed, I architected the first telco Hybris implementation, designing and building the Three UK (H3G) online store. Worth noting is that this was done in a time before Hybris introduced their telco accelerator.

In addition, I have worked for Telefonica O2 evaluating a number of software solutions in an exploratory proof of concept programme.

Having grown to the level of architect working predominantly with Hybris, I have developed a wealth of experience with Hybris itself, integrating it with CRM systems such as Siebel, ERP systems such as SAP, WMS systems such as SideUp and a myriad of other systems such as highly customised reservation management systems in the Travel space, even in one case an RMS system built in FORTRAN.

Leveraging this experience, I am able to consult with customers at a strategic level addressing, often directly with board-level stakeholders. The most common topic at this level is re-platforming the web channel vs a full omni-channel transformation. In my experience while the latter is often the stated goal, most organisations do not really know how to approach it. Hybris is a very capable omni-channel platform, but only if used in the right way.

Once the strategic objectives and approach are clear, I work with the business stakeholders and delivery team to elicit the requirements and design the technical solution to deliver them. This covers everything from data modelling & integration to the sizing and infrastructure requirements. I am in my element thrashing out technical solutions to business problems on a whiteboard with either the business or technical stakeholders – preferably both!

By the time I left Portaltech after over eight years (end of September 2015) as one of three solutions architects my responsibilities included:

• Principal architect of a major on-going programme.
• Principal architect for a number of smaller programmes.
• As arguably one of the most senior technical resources at the company, I consulted on almost all programmes undertaken and guided technical practice and decision-making.
• Coaching technical resources from interns all the way up to those on the architecture track.
• Championing development best practice by introducing development and deployment standards as well as introducing and maintaining tooling such as Sonar.
• Consulting on pre-sales engagements, attending pitches & workshops as well as working with the sales director on responses to RFPs.
• Leading the technical track of the discovery process for new programmes.
• Assisting delivery managers with implementation or team challenges on their programmes.

Services Business Unit Owner

At the end of 2014 I volunteered to take ownership of the Portaltech Reply services business unit, comprising of the operations and application support functions. This role I took on in addition to my existing role as a senior architect. My (self-imposed) objectives were as follows:

• To improve the profitability of the business unit.
• To significantly reduce the number of critical issues and outages occurring.
• To improve the satisfaction of customers with the service being offered.

I take pride in the fact that in addition to achieving all of the above, we also matured the practices of the business unit to a point that we successfully achieved PCI DSS 3.0 compliance. This was no mean feat as the PCI standards are necessarily onerous and rigorously enforced.

As services owner my responsibilities included:

• Ensuring profitability. Managing contracts with customers balanced against resourcing and other costs.
• Coaching the operations lead and his team on standardising and improving practices, simplifying processes and organising a heavy and unpredictable workload.
• Bridging the gap between the delivery teams developing applications and the operations team who then have to host those applications. This involved implementing complementary new practices on both sides.
• Acting as an escalation point for customers and consequently resolving issues when they occurred.
• Coaching the 3rd line development team on diagnosing and resolving issues.
• Managing the PCI DSS 3.0 compliance programme of change as well as the audit itself.

Working in and being ultimately responsible for the services side of the business was very challenging. Consequently however I learned a huge amount. Indeed, my experience here better equips me in addressing issues of infrastructure, hosting and support during all stages of a programme lifecycle and as such, makes me a better solutions architect. While I am not sure if I would do it again, it is experience that no other architects I know can draw on and as such I am glad to have done it.

The tangible skills I developed/reinforced in this role include:

• Deploying applications in cloud infrastructure, specifically AWS and Google Cloud platforms.
• Advanced performance and resilience tuning of JEE application stacks – Load balancer, webserver, application and database. Specifically connection pool handling and configuration.
• PCI DSS 3.0 compliance challenges and how to approach them.
• How to properly build and execute a performance test using Apache Jmeter. There is a significant difference between having a Jmeter script and having an effective stress test. Also interpreting the results is not trivial.
• Crisis management. Identifying patterns and transitioning from issue management to problem management.
• How to offer support and hosting services in a way that is achievable and effective. There is no point in an overly onerous support contract that is unachievable.
• Advanced Unix systems administration.