Viewing profile — exAspArk

Hi HN – founder here. We built Telio to power AI agents for 24/7 call and text support. Voice calls and texts need fast access to context. Our approach: aggregate data from multipl…

Hi HN! We're Evgeny and Arjun, and we’re building a simpler way for startups to do data analytics. Since open-sourcing our Postgres read replica optimized for analytics ( https://g…

There are a few different approaches. The main categories, from simplest to most complex: 1) Read replicas with copied data. The most straightforward, allowing using the same SQL s…

Hey HN! We’re Evgeny and Arjun. We’ve built a managed version of BemiDB that syncs with your existing PostgreSQL database and gives you fast analytical queries without heavy ETL pi…

Our plan is to make BemiDB work with dbt by leveraging the Postgres-compatibility (supported dbt adapters https://docs.getdbt.com/docs/trusted-adapters ). So it should be possible …

This is a great DIY setup. We're hoping to compress this stack and simplify it down to a single binary

Yes! BemiDB natively supports two storage layers, a local disk and S3 (we assumed that most people would choose this in production environments to simplify management). When I quer…

Our initial approach was to implement periodic full table re-syncing. We're starting to work on CDC with logical replication for incremental syncing. Here is our roadmap https://gi…

Sorry, we haven't benchmarked it against ClickHouse yet. Our initial point of reference was just Postgres

We haven't tested this with 1TB Postgres databases yet, assuming that most companies operating at this scale already built analytics data pipelines :) I'm curious if you currently …

Our initial approach is to do full table re-syncs periodically. Our next step is to enable incremental data syncing by supporting insert/update/delete according to the Iceberg spec…

That's why our current approach is to build missing or not fully functional features ourselves to move fast. For example, DuckDB performs reads from Iceberg tables not according to…

Great ideas! We'll keep this suggestion related to read/write separation in mind. We started with a simple unified solution, but we'll keep iterating, listening and addressing any …

My few cents: - Compute and storage separation simplifies managing a system making compute "ephemeral" - Compute resources can be scaled separately without worrying about scaling s…

I'd say that querying data from S3 is not ideal when low-latency queries are required. Generally, there could be a few roundtrip requests to fetch metadata (JSON, Avro) and data (P…

Oh, interesting, thanks for sharing it!