Letter from the CTO

Hi, I am Sourabh, the CTO, and Co-Founder at Hevo.

We began our journey at Hevo in mid-2017 with a mission of democratizing access to data for data-driven people in every company. In these few years, we have been able to build a great team and a great product that is loved by 400+ companies across 30+ countries.

In this letter, I have tried to answer some common questions that we come across when we interact with our future colleagues. I would love to interact with you and answer any other questions you might have.

What are we building at Hevo?

We are building a data platform as a service for internet companies. The data platform makes data queryable, accessible, and consumable for business stakeholders in a company.

Traditionally, data systems/apps have been used and operated only by technical people. Hevo is putting this power in the hands of analysts, marketers, and data geeks who are not programmers.

To understand this better, let’s draw a parallel with AWS.

Before AWS became popular, companies used to buy bare-metal machines and install VMs on them. I come from that era, we used to order machines 6 months in advance because our vendor had an SLA of 90 days on providing the hardware!

AWS brought the most basic unit of infrastructure (a compute unit in the form of a VM aka EC2) and democratized its access to everyone through a no-code UI interface. Now you don’t have to be an infra expert to launch your VMs, the size of the ops team went down dramatically as every developer could now launch his/her own VMs, and most importantly it took a few minutes to buy and launch a server as compared to 6 months with bare-metal hardware.

This opened up gates to unlimited innovation. Companies could now grow faster because they didn’t have to over-provision and wait for hardware! AWS itself launched services like RDS, Elastic Cache, ELB, and many more on the back of EC2, thereby democratizing every other infrastructure piece (database, cache, load balancer, storage) for developers. Every other infrastructure as a business (like Atlas, Confluent, Databricks, Twilio) has been possible because of this change.

The same thing is happening with data infrastructure. Data warehouses like Snowflake and BigQuery are equivalent to the EC2 (compute units) in the data space. We are democratizing all other pieces of infrastructure solving problems around Data Pipelines, Reverse Pipelines, Data Modelling, Data Quality, etc.

Data Pipelines is our flagship product, and there is a lot more coming up.

Why is Snowflake (and BigQuery and others) becoming the EC2 of Data space?

It is very important to understand this. Why? Because the growth of the data industry is driven by the adoption of the EC2 of data space. This is happening primarily because of two reasons.

1. Breakdown of Monoliths

How many technology teams do we know of that want to keep monolithic applications? The answer is 0. Monoliths kill velocity and prevent technology teams from scaling. Hence, every product is broken down from a monolith to a multi-service (or microservice) and often a polyglot architecture.

This is great for the technology team. But, what about analytics teams? They don’t like it, because now they don’t have a single database to refer to as a source of truth.

In the absence of application databases acting as a single source of truth, companies have to build another single source of truth. Here comes the data warehouse and the need to set up pipelines moving data from application databases to the warehouse.

Net result, how many technology teams will not set up a data warehouse and will not set up data pipelines? The answer is… 0.

2. Fragmentation of SaaS

Before SaaS came in (in 2010, not so long ago), companies used to buy a large ERP. This ERP (eg SAP) would have modules for each department — Marketing, Sales, Finance, HR, and so on and so forth. If you want to access data from any business application, it is present in this ERP.

With cloud coming up, SaaS became possible. These SaaS companies now didn’t have to ship software on-premise, because they could deploy it on the cloud. This speeded up the innovation cycle and that created superior products than existing ERPs. These superior products allowed business functions to operate with higher efficiency and hence, the business functions started to break away from ERPs to SaaS.

This was great for the business functions. But, what about analytics teams? They don’t like it, because now they don’t have a single application to refer to as a source of truth.

The same story repeated again, net result, what is the number of companies using SaaS that will not set up a data warehouse and will not set up data pipelines? The answer is… 0.

What new opportunities is this change going to generate apart from data pipelines?

If you look closely, for a company the implications of the source of truth moving from an application database or an ERP to a data warehouse are large. It means that all the decisions a company takes will be based on the data in the data warehouse.

Hence, the data in the warehouse has to be:

• Complete
• Accurate
• Fresh
• Annotated
• Discoverable

Also, the companies will perform the following actions on this data:

• Query
• Aggregate
• Monitor
• Export
• Visualize

Each of the above bullet points will give rise to multiple products catering to different audiences in different domains and verticals.

For Hevo, being a frontrunner in this space, this dynamic creates massive opportunities to build and scale products catering to the above problems.

What does technology mean to Hevo?

Technology is at the center stage at Hevo. There are multiple reasons for that

• We are a technology-only company and not a tech-enabled company. What this essentially means is that a tech-enabled company can live with a good enough product, on other hand, for us building a world-class product is existential.
• Our business model is driven by PLG (Product Led Growth). For us, everything revolves around having a great product. The product makes everything else (acquisition, expansion, conversion, and retention) possible.
• The key expertise of our founders is building products. It is said and well known that a company’s DNA is an extension of the founder’s strengths and weaknesses. Our strength is building products, we believe we can challenge anyone in the world on that front.

The proof of the importance of technology is in our financial planning. Any tech-enabled company would spend no more than 5–10% of their funds on technology. We keep aside 60–70% of our funds for building technology. What that means is that our investment in technology would be more than other companies who have raised 10–15 times that of Hevo.

How do we promote learning at Hevo?

The primary act of an engineer is to learn. If you are not learning, you may as well not be working at all.

At Hevo we promote learning for engineers by giving them the freedom to solve problems the right way rather than just making things work.

We also give them the responsibility and opportunity to define the scope and outcome of features/products they are building. Instead of being dependent on product managers to define the specifications, engineers at Hevo are encouraged to drive product direction themselves.

We also give them opportunities to move around into different teams for exponential learning.

How is work assigned at Hevo and what will I be doing?

We do not confuse agility with chaos. We don’t believe in breaking things when things can be built without breaking stuff. At Hevo, we follow a structured and data-driven roadmap planning approach. At any point in time, an engineer knows what he/she is going to work on for the next few months.

Here is how:

• The roadmap is planned every quarter at a collective level. What goes into the roadmap is based on data and customer feedback that we have received in the past few months.
• The roadmap is broken down into product and engineering goals.
• Each engineer is given a MIG (Massive Impact Goal) for the quarter. The engineer is expected to spend 60–70% of his/her bandwidth on this Goal.

What is the technology stack? Are there any unknowns that I should be aware of?

At Hevo, we follow the KISS principle. We don’t overcomplicate things, all technology decisions are discussed and deliberated. We are built on well known open source stack — Java, Kafka, MySQL, Redis, RocksDB, React, and Angular.

What is our product roadmap?

I won’t bore you by listing big jargon product features or services that don’t and will never exist. We think about the roadmap as a journey and not as a checklist.

Our product roadmap is based on four pillars:

User Experience

We want to provide a seamless, effortless, and error-free experience to our users. We obsess over the smallest of the details and are always willing to go the extra mile to deliver an awe-generating experience.

This is, however, much easier said than done. User experience is not just about the UI. When you experience the thud of a Tesla door, or the roundness of your iPhone’s edge, or a dancing icon in a google’s app all you are witnessing is a team’s tireless efforts in making things perfect. User experience is about building products that just work and about automating the simplest of steps the user has to perform to get the job done.

As an engineer at Hevo, you will get to live this journey of building perfect products.

Connectivity

Connecting to other applications is core to what we do. We are an iPaaS (Integration Platform as a Service) and hence, a user can only benefit from our technology when they can integrate their systems using Hevo. We currently connect to more than 70 databases, applications, and file storage services. We want to expand our connectivity to more than 200 such systems by next year.

Performance

We process humongous amounts of data. At the moment we are processing close to 200 Billion records every month through our systems. Our systems at peak clock up to 200K records/second. We are at a scale where adding an extra line of log creates a huge difference in performance. And, we are just getting started. The scale will go up to a trillion records in a few months and will keep growing exponentially.

Resilience

To explain what resilience means to us, I will go back to our initial comparison with AWS. Imagine if EC2 goes down every now and then, what will it do to your services? That is the same impact our users would feel if there is a downtime on Hevo.

That doesn’t mean services don’t go down at Hevo or APIs don’t fail, they do. However, we design our systems to be resilient to these failures, and that shows up on our status page.

Additional benefits of working at Hevo

There are many benefits of working at Hevo. To list some of them:

• A true startup experience, minus the chaos and instability. We are very well capitalized, our runway would last a decade because our customers pay us handsomely. We are organized in thought process and take extra care of keeping things in order.
• Work with the fastest-growing company in the data space from India.
• A world-class, driven team to work with.
• Get to solve problems in depth rather than just making things work.
• We promise good career growth.
• A trust-driven and open culture where your views are respected irrespective of your years of experience or industry background. Check out one of our colleagues talking about that here.
• We take care of our employees.
• Everyone is helpful at Hevo. We only compete with our competitors, we don’t compete with each other.

I hope I have given you a little more clarity on what it is like to be a part of Hevo. We are building a team of 100 outstanding Developers, QAs, Product Managers, and Designers at Hevo Data.

The world of data is changing and is never going to be the same, if you think it is worth being a part of our mission, give us a buzz at 100guns@hevodata.com.