DevOps is receiving considerable focus as a growing number of organizations adopt more advanced tools to assist developers in managing increasingly intricate systems and workloads. Recently, Databand – an AI-powered observability platform designed for data pipelines, specifically to pinpoint issues with data sources when engineers utilize a variety of data management tools – successfully completed a funding round of $14.5 million.

Josh Benamram, the CEO and co-founder of the company alongside Victor Shafran and Evgeny Shulman, stated that Databand’s plans encompass increased recruitment; continued acquisition of customers for its current offering; expansion of the range of tools available to users, covering the expanding landscape of DevOps software, with a strong commitment to open-source resources; and investment in the future development of its commercial product. This future development will include automated problem resolution, enabling engineers to not only identify issues but also begin automatically correcting them.

This Series A funding round is led by Accel, with contributions from Blumberg Capital, Lerer Hippeau, Ubiquity Ventures, Differential Ventures and Bessemer Venture Partners. Blumberg Capital previously led the company’s seed funding round in 2018. To date, Databand has raised approximately $18.5 million, but has not disclosed its valuation.

The challenge that Databand addresses is becoming increasingly critical and widespread (as demonstrated by the rapid, exponential growth in globally stored zettabytes of data each year). As data workloads continue to expand in both size and application, they simultaneously grow in complexity.

Furthermore, organizations typically employ a diverse array of applications and platforms for managing source data, storage, and utilization. Consequently, identifying the source and nature of glitches within any single data source can be difficult. Manual troubleshooting can be extremely time-consuming, and in some cases, impossible.

“Our customers were consistently facing difficulties with ETL (extract, transform, load) processes,” explained Benamram, speaking from New York (the company maintains offices in New York and Tel Aviv, with development and operations teams also located in Kiev). “Users lacked a clear method for organizing their tools and systems to ensure the production of dependable data products.”

He emphasized the difficulty of prioritizing failure analysis when engineers are simultaneously managing analytics dashboards, monitoring machine learning model performance, and addressing other demands. This is further complicated by the possibility of data suppliers altering APIs, potentially disrupting the data source entirely.

Benamram noted that receiving inaccurate data can be highly frustrating, and potentially damaging. He stated that anomalies are often overlooked by engineers until brought to their attention by executives observing discrepancies in their dashboards – a less than ideal situation.

Databand utilizes big data techniques to effectively manage large datasets. It analyzes various data points, including pipeline metadata such as logs, runtime information, and data profiles, alongside data from platforms like Airflow, Spark, and Snowflake. This information is consolidated into a unified platform, providing engineers with a comprehensive view of system status, enabling them to identify bottlenecks, detect anomalies, and understand their root causes.

Several companies are developing data observability tools – Splunk is a prominent example, alongside smaller companies like Thundra and Rivery. These companies may expand into the specific area that Databand is addressing, but Databand’s dedicated focus on identifying and resolving anomalies has established a strong market presence.

Accel partner Seth Pierrepont explained that Databand initially gained Accel’s attention through a practical demonstration of its value: Accel itself required a solution like it for its internal operations.

“Our internal data team at Accel was encountering challenges with data pipeline observability. Even at our relatively modest scale, we experienced weekly issues with the reliability of our data outputs, and our team discovered Databand as a solution,” he said. “As organizations across all sectors strive to become more data-driven, Databand provides a crucial product that guarantees the dependable delivery of high-quality data. Josh, Victor and Evgeny possess extensive experience in this field, and we have been impressed by their collaborative and insightful approach to assisting data engineers in managing their data pipelines with Databand.”

The company’s solutions are utilized by data teams ranging from large Fortune 500 corporations to smaller startup companies.