Sometimes do things that DO SCALE after all!

Hello — I am a software solutions architect at Gyana. A foundational member- I have seen it being conceived, born and now scaled.

I have worked earlier with top MNC’s like Cisco & Juniper, in both big and small teams. I have seen that side of the world. My self-start journey started with a non-profit org named Anahatalife. Now, I am on another marathon journey with like minded and enthusiastic people in Gyana. It’s like a team who has to win a relay race but to do so every individual has to run as per his max potential. It wouldn’t be wrong to say there we need to believe there are NO LIMITS!

“Limits are imaginations from past or historical records.”

[I am also an avid marathon runner — therefore from time to time you will find I cannot help but keep bringing up running metaphors.]

History always teaches you to be someone other than you. History is safe. History is limited.

And that’s why History is HISTORY!

Be bold.

Be limitless.

Be maker of your own destiny.

Create unprecedented space for your endeavours.

Now as we are scaling up Gyana, we want to rewrite our own history- as someone wise said — “the best competition is always with your own former self”.Thankfully we have put in place processes and infrastructure solutions over the last 2 years which otherwise I would expect only in some big organisations or sometimes even not. This has helped us begin to scale up quite fast and with zero casualties!

I know most startup books say “do things that do not scale” — but myself ( and Founders ) always strongly believed that ‘history’ is limiting in nature.

There are no golden rules when it comes to startups.

We knew that our entire business potential is such that scale up is not just a essential step in the business- the nature of our work is such that it is the ONLY step that will ultimately make sense. So, we constantly worked towards band aid solutions as well as long term remedies- be it cyber security, ISO certification very early on, massive ETL pipeline automations, sorting algorithms, classifiers, orchestrators, data lake strategies — CTO David Kell always ensured we are scale ready from very early on.

And how it paid off ! As soon as we begun the customer ramp up process — I can tell you how much we thanked each other for that foresight. I pat myself on the back quite often for going that extra mile each time in the past architectural plans.

“What seemed like more work turned out to be a pure blessing.”

Some tips from my end :-

1. Agile methodology — set key processes

I have seen big organisations struggling to adopt Agile methodology even though they want to. The data science market is very competitive and businesses want faster feature delivery to their customers. Agile enforces us to have key processes in place which helps in quicker delivery along with quality. It is definitely helping us to have high feature velocity.

“Adopt it early and keep scaling it as you grow.”

2. Code quality measures — follow industry standard best practices

This is always key concern in s/w development. You don’t want to waste your bandwidth on fixing more issues than development. More importantly you don’t want to lose your customer’s trust. We strictly follow industry standard code reviews, coding standards, versioning system, documentation etc. This is definitely helping new on-boarding members as we scale up. Beautifully documented code is a dream- always has been — and we got very close to it. Believe it or not- I and CTO do spot checks on code documentation even now, for everyone. On some rainy days , our CEO — an ex-techie herself, still drops that occasional slack message to a random coder “I checked the codebase, your modules are not well documented” even now. [ Sometimes I think she has made clones of herself].

3. Get the Certifications early

“If your customer cares about trust, then so shall you give.”

We are also ISO27001 certified (framework of policies and procedures that includes all legal, physical and technical controls involved in an organisation’s information risk management processes). Many people said it is an overkill but given we work in enterprise Saas, this alone changed our pipeline by 30% overnight.

[So don’t listen to people- only you know the business best.]

4. Continuous Integration & Deployment — process optimisation

With high feature velocity, it takes a lot of effort and bandwidth to test and deploy newer version of code. Hence, in order to optimise this overall process it is important to deploy continuous integration and deployment. I am happy to say I have not seen this level of 100% autonomous deployment in very established organisations.

5. Analytics — User behaviour, product monitoring/logging

It’s paramount to have a feedback loop which helps to improve product by every release. Direct user feedback is always great. However, user will not always be candid, will not always have time and will not always talk to you directly ( will talk to others in team or worse, to competitors ). Deep stack analytics are the plan to have alternate infrastructure in place with which you can understand the improvement areas in your product. We have developed in-house solutions advanced solutions to monitor our product performance and hence we can constantly improve it regardless of user feedback

[However, my advice would be to use these services smartly. Don’t get too coupled with proprietary services that it would be difficult to switch to any other service provider in future]

All in all keep a balance — startups are all about fast response and delivering quick. So keep doing what you do but also keep an eye on long term strategies. Ear mark that Saturday morning or that one unusually easy day for cleaning up and creating long term grooves in your product architecture and deployment practices. Make an organization that does not stumble and break when you go from local to global. Create a culture that is all about go-getters but also about sustainability and far sightedness. I am glad I found that in Gyana foundations from day one.

I would love to hear from fellow readers about their journey from startup to scale-up mode. How did you achieve this?