Cloud adoption was one of the most disruptive technology trends in the last decade. The emergence of AWS, Microsoft, Google & other cloud infrastructure providers, and the adoption of cloud by large enterprises — especially players in the tech ecosystem like Netflix & Uber — gave birth to an entire industry of cloud & application monitoring solutions. Companies like Datadog, NewRelic, & Splunk launched to help companies better monitor their cloud & application infrastructures & manage their web application performance in real-time.

Today, both the cloud infrastructure stack & the amount of data processed on it have flourished compared to a decade ago.

• Around 50% of all corporate data is stored in the cloud,
• infrastructure has evolved (e.g. containers & serverless arrived), and
• an enterprise processes 40–50x the amount of data on a daily basis than it did in 2010.

The abundance & the velocity of data, coupled with complex infrastructures becoming ever more sophisticated every year, resulted in the modern event-driven cloud & application monitoring stack (see below), requiring companies to use multiple service providers in verticals such as data streaming, architecture / data lakes, event correlation, modelling & model monitoring, labelling, investigation and event handling to manage their workflows.

This “observe-detect-act” flow necessitates organizations to invest in expensive ML stacks with multiple tools that create time consuming and burdensome processes. And even with such an investment, the efficacy of the system deteriorates over time, and requires teams of engineers who have to keep track of these microservices’ integrations to make sure they perform.

Synnada (named after an ancient town in Turkey) is targeting this problem with a novel approach: It delivers a self-contained ML model that discovers anomalies from a complex data stream, helps users generate accurate alertsand continuously monitors the health of these alerts by learning from user feedback. The platform creates a continuous optimization process that discovers alerts with unsupervised learning, investigates edge cases to find mission-critical alerts & monitors modelling for accuracy. Thus, Synnada enables users to build observe-detect-act applications on complex streams in a low-cost manner by converging labelling, alert discovery & investigation steps.

Co-founders Ozan & Samican are a great fit to tackle this problem.

• Ozan is an AI scientist with solid academic (ODTU, Stanford) and professional (Optumsoft, Striim, and Facebook) backgrounds and built similar systems in media, finance, industrial IoT verticals.
• I’ve known Samican for over 7 years now; we met back when he was the Head of Operations @ Apsiyon. In that time, he grew teams from 20 to 100 people & helped raise $35M+ in funding for Apsiyon & Picus. He also co-hosts the leading local podcast on entrepreneurship in Turkey.

So it is no surprise that we are very excited to lead Synnada’s pre-seed round & to work with Ozan & Samican to elevate the modern event-driven application stack.

‍