Hacker News: buremba

New comment by buremba in "Building durable workflows on Postgres"

buremba — Thu, 28 May 2026 23:56:09 +0000

Thanks for correcting me, I was wondering why I got downvoted. :)

Clickhouse is also great for OLAP for sure

New comment by buremba in "Building durable workflows on Postgres"

buremba — Thu, 28 May 2026 21:37:36 +0000

Ahh, sorry about that. It should be fixed in an hour, looks like we mixed the permissions. I just tried and confirmed other login methods work if you would like to try out.

New comment by buremba in "Just Use Postgres for Durable Workflows"

buremba — Thu, 28 May 2026 20:05:31 +0000

The native extensions are fine but I don't have good experience with any third party extensions, so far tried Timescale, pg_lake, citus, and pgvectorscale. They look very appealing but it's usually a trap as you can't get the value without using the vendor's cloud offerings.

I think if you grow enough to look for these extensions, it's usually better to bet on purpose-specific tooling. For example, I use DuckDB/Iceberg combination extensively for columnar data and connect DuckDB to PG when I need it.

New comment by buremba in "Just Use Postgres for Durable Workflows"

buremba — Thu, 28 May 2026 20:00:51 +0000

Yeah I have logs in Sentry, which also uses Postgresql.

New comment by buremba in "Building durable workflows on Postgres"

buremba — Thu, 28 May 2026 19:39:49 +0000

All you need is Postgres until you scale into TBs of data. We use Postgresql as a durable workflow engine, vector search, time-series data, BM25 search, OLTP/OLAP engine, and a queue. It's basically the only dependency we have for https://lobu.ai

The main benefit is centralizing all the data in one place so we don't need to worry about copying data in between multiple systems. Once something becomes the bottleneck, you can eventually migrate to a purpose specific tool to scale out.To be honest, LISTEN/NOTIFY in my opinion is the most fragile part of PG but it's fine as start until you scale out.

New comment by buremba in "Cursor Camp"

buremba — Wed, 29 Apr 2026 19:57:55 +0000

I don't usually like my cursor to be hijacked but man, I like this one.

New comment by buremba in "Show HN: Ctx – a /resume that works across Claude Code and Codex"

buremba — Tue, 21 Apr 2026 18:27:53 +0000

Yes that's it. I actually just ask codex/claude code to look up the session id when I want to resume sessions cross harness, it's just jsonl files locally so it can access the full conversation history when needed.

New comment by buremba in "Show HN: Ctx – a /resume that works across Claude Code and Codex"

buremba — Tue, 21 Apr 2026 16:57:12 +0000

Since prompt caching won't work across different models, how is this approach better than dropping a PR for the other harnesses to review?

New comment by buremba in "Ask HN: What Are You Working On? (April 2026)"

buremba — Sun, 12 Apr 2026 22:53:41 +0000

I'm working on multi-tenant version of OpenClaw for organizations that has shared memory layer. It includes an entity based agent context layer that can be used as OpenClaw plugin and a sandbox runtime layer which uses just-bash with pi and let you expose the context via a bot an API.

https://lobu.ai

New comment by buremba in "OpenAI says its new model GPT-2 is too dangerous to release (2019)"

buremba — Wed, 08 Apr 2026 03:05:01 +0000

It playbook is that a model is too dangerous until a competitor releases a competing model that beats yours.

New comment by buremba in "Show HN: Mtproto.zig – High-performance Telegram proxy with DPI evasion"

buremba — Sat, 04 Apr 2026 03:10:34 +0000

You can ask your agent to verify or review code. Just because people wrote code by hand, it doesn't mean you should trust

New comment by buremba in "Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw"

buremba — Sat, 04 Apr 2026 00:47:25 +0000

I get why they block OpenClaw and it makes sense but I wonder if they can actually detect OpenClaw calling Claude Code CLI using something like acpx.

It's simply identical to how people use Claude Code locally.

New comment by buremba in "Show HN: Mtproto.zig – High-performance Telegram proxy with DPI evasion"

buremba — Sat, 04 Apr 2026 00:31:06 +0000

does it matter?

New comment by buremba in "Delve allegedly forked an open-source tool and sold it as its own"

buremba — Thu, 02 Apr 2026 17:35:59 +0000

Compliance tech company who doesn't know about open-source. Interesting.

New comment by buremba in "MCP Is Overengineered, Skills Are Too Primitive"

buremba — Thu, 02 Apr 2026 16:03:45 +0000

I extensively use uv (IMO better than venv) but still it's Python specific and not universal. npm is much worse and native binaries are almost impossible to deal with multiple versions.

nix is specifically targeting this use-case and it'e extensively used by vendors like Replit.

New comment by buremba in "MCP Is Overengineered, Skills Are Too Primitive"

buremba — Thu, 02 Apr 2026 15:33:06 +0000

They're composable but computers are not. Two skills might depend on a different version of a dependency which is pretty hard to maintain and their needs to be a deterministic system (agents are not) to resolve the conflicts and make sure two skills can live in the same environment.

New comment by buremba in "MCP Is Overengineered, Skills Are Too Primitive"

buremba — Thu, 02 Apr 2026 15:30:41 +0000

I vote for taste

MCP Is Overengineered, Skills Are Too Primitive

buremba — Thu, 02 Apr 2026 14:42:54 +0000

Article URL: https://lobu.ai/blog/mcp-is-overengineered-skills-are-too-primitive/

Comments URL: https://news.ycombinator.com/item?id=47615192

Points: 3

# Comments: 6

New comment by buremba in "When does MCP make sense vs CLI?"

buremba — Mon, 02 Mar 2026 12:19:51 +0000

CLI enables the actions to be made on behalf of you, the external service is not aware whether it's you or AI making the calls. With MCP, Sentry knows it's AI making the call so can be smarter about the security. There is many MCP annotation hints on tools to mark the as destructive, read-only etc.

New comment by buremba in "When does MCP make sense vs CLI?"

buremba — Sun, 01 Mar 2026 21:52:03 +0000

There is nothing wrong with MCP, it's just that stdio MCP was overengineered.

MCP's Streamable HTTP with OAuth discovery is the best way to ship AI integration with your product nowadays. CLIs require sandboxing, doesn't handle auth in a standard way and it doesn't integrate to ChatGPT or Claude.

Look at Sentry, they just ship a single URL https://mcp.sentry.dev/mcp and you don't need anything else. All agents that supports MCP lets you click a link to login to Sentry and they make calls to Sentry to fetch authentificated data.

The main problem with MCP is the implementation. Instead of using bash to call MCP, agents are designed to make single MCP tool calling which doesn't allow composability. We solve this problem with exposing MCP tools as HTTP endpoints and it works like charm.