The Evolutionary Journey of Middleware with a Spotlight on Observability With Laduram Vishnoi

In today’s fast-paced digital landscape, ensuring the smooth operation of complex systems is paramount. That’s where Middleware steps in. As a full-stack cloud observability platform, Middleware is dedicated to revolutionizing the way development and operations teams monitor and manage their environments.

Middleware is the ultimate solution for cloud observability. Engineered to streamline monitoring, it consolidates metrics, logs, traces, and events onto a unified timeline for swift issue resolution. Leveraging advanced AI, Middleware provides insights into infrastructure and application performance, empowering proactive optimization. Welcome to Middleware: Where observability meets innovation, propelling businesses forward with confidence.

Laduram Vishnoi is the Founder & CEO of Middleware, a trailblazer in observability. Formerly, he led Acquire.io from its inception to a team of 200, raising $57M in funding. A seasoned angel investor and advisor, Laduram’s journey spans Jodhpur, London, and San Francisco since 2016, infusing global perspectives into his work, driving innovation, and fostering growth.

The interview delves into Middleware’s journey, highlighting its focus on observability within the AI landscape. Amid discussions on entrepreneurship and AI innovation, Middleware’s strategic approach to observability emerges, emphasizing its critical role in optimizing performance and driving continuous improvement in AI-driven solutions. This focus underscores Middleware’s commitment to delivering impactful and reliable AI services while navigating the complexities of the tech startup ecosystem.

AIM: How has your journey been in San Francisco, and what aspects of the city do you particularly enjoy? Could you elaborate on why you chose to build your business specifically in San Francisco, and what experiences have shaped your journey there?

Laduram Vishnoi: I’ve been in San Francisco for seven years now. I moved here in 2016 from London. I started a company called Acquire in London in 2015, and we were getting quite a few customers from the U.S. So, I decided to try it out in San Francisco. I love the ecosystem, and it’s built around software companies like SaaS companies. I just stayed here; I sold everything and made San Francisco my permanent home. It was hard at first. I ended up living in a horrible area in San Francisco because that’s where you get the cheapest apartments. I stayed there for almost two years. We couldn’t raise the money early on but progressively did many things.

AIM: What is it about the energy and atmosphere of cities like San Francisco, particularly in tech hubs like the Silicon Valley, that fosters such a strong desire among individuals to start and scale their own ideas, even when working for large corporations with lucrative paychecks?

Laduram Vishnoi: To be honest, I feel that when I return from London or India there is a difference in the startup ecosystem. It seems much easier to start something in San Francisco. When you share ideas with friends and family there, they’re genuinely enthusiastic and willing to support. They might even offer to join you in the venture. The level of excitement and encouragement from the community in San Francisco is much higher. However, when I talk to my friend in London, the response is quite different. While they acknowledge the potential of the idea, they tend to focus on the complexities and challenges, often suggesting a delayed start by three or four years. But in San Francisco it’s “Just go and do it.”

AIM: What fuels the continuous pursuit of entrepreneurship, even among individuals with stable jobs in large corporations, particularly in cities like San Francisco? How does the city’s vibrant atmosphere and culture of innovation inspire and facilitate the creation of startups like Middleware?

Laduram Vishnoi: So, the thought process was like this: I left my previous company, took a break, and then dabbled in full-time investing, which I didn’t enjoy. I found myself at a crossroads: either return to my previous company or start something new. The spark of the idea came around 2017-2018 when we were building my previous company, we were looking for observability products and we were using cloud, microservices setup and all the latest technology. We were looking at observability as a monetary product. Surprisingly, there weren’t many options available. One company provided a quote for observability, but their pricing was three times higher than our actual cloud bill, which was around $4,000 a month. It was a shock, and we couldn’t afford it. I casually mentioned to other engineers in my team that we could build a smaller version of observability, similar to what we had in my previous company. So, when I started the new company, I hired 15 engineers from day one. For almost a year, we quietly built the software without engaging with VCs or customers. That’s how it all began.

AIM: Can you discuss the unique selling point that differentiated Middleware in the cloud observability ML market, and how it impacted your initial sales strategy? Could you walk us through the experience of acquiring your first customers and how your proposition resonated with them?

Laduram Vishnoi: I’d say our first sale came through inbound efforts. I’ve heavily focused on inbound strategies, leveraging my networks here in Silicon Valley, where I’ve lived for seven years and raised over $60 million in venture funding. Securing a substantial first customer was crucial before diving into website optimization. Our primary focus is on SEO and content, not just about our product but raising awareness about our presence in the industry. It’s about understanding the gap in observability and how we differentiate ourselves from others. The first customer came inbound, signed up, entered their credit card details, and it was a pivotal moment for us. Embracing a product-led growth strategy, we have no sales team. Our customer success and support teams are heavily focused on PLG, with features tailored to enhance user experiences. While I occasionally get involved in dealing with customers personally, our focus remains on PLG.

AIM: How can technology companies in AI and data science effectively build brand recall value and establish a strong media presence without resorting to aggressive sales tactics? Could you discuss strategies such as community engagement, participation in conferences, and digital platform utilization to foster brand credibility and resonance within the industry?

Laduram Vishnoi: It’s become expensive. You can indeed build a brand using a sales team, having a large number of them engaging in outreach. Personally, I don’t consider myself a brand or media expert, however, I feel strongly that you can build a brand through software, in Devspace or Infraspace where we are and where you provide value for others. Helping others always tends to come back to benefit you. By fostering communities and continuously offering assistance, you naturally attract people to your brand. This is something we actively practice here. For instance, we frequently organize events where notable speakers share insights. Recently, the founder of Post Grad spoke at one of our events, drawing in nearly 50-60 engineers. We didn’t ask for anything in return, yet we received a significant number of signups afterward. I believe in leveraging events and initiatives like these to build a brand effectively without incurring significant costs.

AIM: What was your approach to building a successful product idea in response to the challenges posed by expensive cloud observability platforms? Could you discuss the factors and frameworks you considered to ensure the scalability and success of your product, while prioritizing customer success and delivering on your promises?

Laduram Vishnoi: From day one, we understood the scale we were dealing with. Our customers would be sending petabytes of data, potentially reaching peak volumes. So, we knew we had to build our product to handle such massive amounts of data. That’s why it took us almost a year and a half to establish a solid foundation for the product. We aimed to ensure that we could handle 10 to 20 petabytes of data across all platforms per month and scale accordingly. There were challenges early on, but as we iterated, we identified numerous gaps in the market, particularly concerning new technologies like microservices, and cloud-native and the observability in those parts. These environments generate vast amounts of data, and simply dumping it into a data lake or ocean makes it incredibly difficult to manage for engineers. How would you give great data that they can actually use?  We recognized the need to provide actionable insights from this data overload. Hence, we leveraged AI to analyze and debug the data for DevOps teams. By training our algorithms, we enabled significant improvements and bridged the gaps. While pricing may not be the primary differentiator for us compared to legacy companies, AI plays a crucial role. However, it’s essential to note that AI isn’t our sole observability; we offer complete observability, covering everything from logs to metrics, traces, and events. AI simply fine-tunes this data to provide maximum value to our customers.

AIM: How do you navigate the dilemma of whether to hire data professionals first or secure clients first when starting a data science services venture? Considering the financial constraints and the challenge of scaling the team, how do you address the “chicken and egg” problem of balancing talent acquisition with client acquisition?

Laduram Vishnoi: In business, customers always come first. But if you don’t have a product, what can customers do? So, the first step is to build a team and a product. Once your product is stable and scaling, then you shift your focus from the team to the customers. Initially, you need to focus on building an amazing team and an amazing product from day one. Then, once you believe you have a solid product, perhaps after releasing a better version, the priority becomes the customers. 

For the financial gap is when venture capitalists (VCs) come in. They fund companies to fill the gap between building and selling to customers because they don’t have the resources to do both. So, VCs play a crucial role in bridging that gap.

AIM: How did you strategically network to secure a meeting with a VC before pitching your idea to them? Can you elaborate on your thought process leading up to the decision to approach VCs and how you prepared for the meeting?

Laduram Vishnoi: So the first thing is, having a decent network here is advantageous. I personally know many VCs here. But what about those who don’t have those connections? It’s tough. It’s really tough. The easiest way for them is to join accelerator programs. These programs bring people together, providing opportunities for networking. For instance, we had middleware, and we went through Y Combinator, which is an amazing accelerator program. They have over 10,000 VCs on their portal. When you launch your startup publicly through Y Combinator, you’ll receive hundreds, even thousands of requests from VCs wanting to talk to you. However, getting into these communities is highly competitive. They might have 30,000 applicants but select only 200. It’s challenging, but there are many other excellent accelerator programs out there to try. From my personal experience, in my previous startup, I didn’t have networks. We applied to YC four times but didn’t get in. So, we joined another accelerator program called 500 Startups. That’s where we built our entire network. It took us two years of hard work, attending various events and meetings, to establish those connections. Eventually, we managed to raise the funds we needed.

AIM: What is the one crucial aspect that matters most to VCs during a pitch, especially in the limited time frame of about half an hour?

Laduram Vishnoi: We’re discussing the very early stages of a company. When it is pre seed. But at this point, for VCs, it’s all about the team: who are the founders, what is their experience? Have they successfully built similar things elsewhere? Or are they newcomers, perhaps transitioning from other careers? Personally, I believe the idea isn’t as crucial at this stage. Ideas can always evolve, but having a strong team is paramount. A competent team can execute faster, adapt the idea, learn quickly, and make the right decisions. Conversely, if the team isn’t strong, they might cling to a bad idea for too long, burning through resources in the process. That’s the key distinction.

It’s incredibly challenging to build something from scratch because, quite simply, you lack the experience initially. You have to learn as you go, and for founders, this can be particularly tough. They often find themselves stuck, with egos getting in the way. Stubbornness and ego can be major obstacles. When someone advises them to do something differently, they might brush it off, insisting they know best. They’ll attempt it their way, fail, and only then realize their mistake. But by then, several months might have passed, turning into a nightmare. This cycle often repeats itself for six to nine months, creating a challenging journey for many founders.

AIM: How did you sell your idea of Observability?

Laduram Vishnoi:I believe observability has been in the market for about a decade now, spanning three different generations. Initially, we had on-premises solutions, where companies managed their own data centers either in their garage or office. This was the first generation. Generation two started around 2008-2010. Companies like Azure and AWS emerged during this time, primarily focusing on monitoring monolithic applications and traditional cloud environments.

However, around 2018, with the rise of kubernetes changed the game for observability. Microservices and cloud-native architectures, the landscape began to change. Companies needed solutions that could handle auto-scaling, containerization, and dynamic environments. This shift marked the beginning of the third generation of observability.

Middleware entered the scene during this third generation, showcasing the evolution from the second generation’s efficiency to a more sophisticated approach. While traditional monitoring tools focused on service monitoring in an outdated manner, middleware concentrated on adapting to the new challenges posed by modern technologies like Kubernetes.

The volume of data being produced has increased exponentially over the years. Approximately 95% of the data on the internet has been generated in the last five years alone. With this surge in data, traditional methods of monitoring have become inadequate and expensive. Therefore, there is a need for scalable technologies and advanced debugging solutions tailored to modern environments like Kubernetes.

This is where Middleware stands out. We are entirely focused on the latest technologies and are empowered by AI to enhance observability further. So, false positives are a common occurrence in observability. Typically, an observability system might send you 100 notifications a day, but only a small portion, say 20 or 30, may actually indicate genuine issues, while the rest are false positives. However, with AI, we’ve made significant strides in distinguishing between false positives and genuine alerts. AI algorithms are adept at recognizing patterns and anomalies, allowing us to filter out false positives more effectively. This capability has been instrumental in improving the accuracy and efficiency of our observability platform. AI helps us differentiate between genuine issues and false positives, significantly improving the efficiency of our monitoring and debugging processes.

AIM: What are some of the use cases that are driving observability in the realm of AI, particularly with regards to Generative AI?

Laduram Vishnoi: Gen AI in the realm of observability is still in its early stages, but it holds immense potential. One significant area where Gen AI is making strides is in filtering and analyzing data to provide actionable insights. For example, imagine a scenario where hundreds of errors and alerts are being generated by your infrastructure. By leveraging Gen AI, we can sift through this data and pinpoint the actual issues while disregarding false positives. This allows us to provide the DevOps team with accurate and relevant information, enabling them to address issues promptly.

Moreover, AI can learn from these incidents and proactively alert the engineering or DevOps team about potential issues before they escalate. This proactive approach to problem-solving is crucial for maintaining system reliability and minimizing downtime. In essence, AI is revolutionizing observability by not only identifying issues but also anticipating and preventing them, ultimately making a significant difference in the effectiveness of DevOps teams.

AIM: What are your visions for the growth of the industry in the next one year, five years, and ten years, particularly in the evolving fields of AI and data science? How do you plan to stay ahead and adapt to the ongoing evolution of the industry?

Laduram Vishnoi: So, we currently have a team of 30 people working out of our own office. Our focus for the next year is clear: it’s all about go-to-market strategies and acquiring more customers. Simultaneously, we’re looking to scale our team. While our core tech team is primarily based there, we also have some engineers working remotely. But we are going to scale this year.

AIM: When it comes to raising money for startups, how does Middleware ensure efficient utilization of funds? 

Laduram Vishnoi: In the early stages, when you raise money, it’s crucial to handle it with extreme care because it’s not your money. You can’t afford to be reckless with your spending. Every dollar needs to be invested wisely, with a clear understanding of the potential outcomes. For example, if we raise money and decide to invest it in marketing initiatives like attending conferences and setting up booths, we need to carefully evaluate the outcomes, whether they’re physical or non-physical. While the funds may not come directly from our own pockets but rather from investors, it’s crucial to be prudent in our spending. Even if the expenses are covered by customer payments and contribute to our revenue, we must remain mindful of the fact that it’s ultimately the investors’ money we’re utilizing. Therefore, it’s essential to exercise caution and ensure that every decision, no matter how small, is made with care and consideration.