News for hackers. Stuff that matters.

nishantmodak

nishantmodak on Hacker News

Re:Call a Human MCP on Monday March 02, 2026 @07:31PM (#47226294)
I built a small MCP server that lets any AI agent (Claude Desktop, Cursor, Windsurf) block and wait for a human response before taking irreversible actions.

  Two tools:
  - ask_human(question) – agent blocks until you reply
  - request_approval(action) – agent blocks until you click Approve or Deny

  Notifications via Slack (with interactive buttons), or native macOS
  dialogs. Works with any MCP-compatible client over stdio or SSE.

  The use case that motivated it: I kept coming back to find my agent had done
  something I didn't want while I was away. Now it just asks.

  GitHub: https://github.com/nishantmodak/call-a-human-mcp
Re:I built Vector. Now I'm answering the question your observability v... on Wednesday January 14, 2026 @06:09PM (#46625302)
Problem has never been the storage. Its running those queries to return in milliseconds - if its for a dashboard, an alert of your new AI agent trying to make sense of it.
Re:Why high-cardinality labels stress metrics databases on Thursday January 01, 2026 @03:57PM (#46457933)
I wrote this after seeing the same failure modes across multiple teams and metrics stacks.

The intent here isn’t to argue that high cardinality is bad but to explain the mechanical reasons unbounded cardinality shows up later as cost surprises, slow queries, and trust erosion.

Curious if others have seen similar query-time or indexing issues they have noticed.

Re:I am rich and have no idea what to do on Friday January 03, 2025 @04:23AM (#42584042)
Wasn't it pascal who said that all of our problems come from our inability to be bored.
Re:Why is observability so expensive? on Wednesday April 03, 2024 @05:41PM (#39923575)
I have a different take

>Engineers have to pre-define and send all telemetry data they might need – since it’s so difficult to make changes after the fact – regardless of the percentage chance of the actual need.

YES. Let them send all the data. The best place to solve for it is at Ingestion.

There's typically 5 different stages to this process.

Instrumentation -> Ingestion -> Storage -> Query (Dashboard) -> Query (Alerting)

Instrumentation is the wrong place to solve this.

Ingestion - Build pipelines that allow to process this data and provide for tools like streaming aggregation, cardinality controls that allow to 'process it' or act on anomalous patterns. This atleast makes working on observability data 'dynamic' instead of having to go change instrumentation always. Storage - Provide blaze (2hours), hot(1 month), cold(13 months) of tiered data storage with indipendent read paths.

This, in my opinion has solved for the bulk of cost & re-work challenges associated with telemetry data.

I believe, Observability is the Big Data of today, without the Big Data tools! (Disclosure: I work at Last9.io and we have taken a similar approach to solve for these challenges)

https://last9.io/data-tiering/

Re:Clever Hans Fallacy of Monitoring Tools on Wednesday July 29, 2020 @03:04AM (#23984130)
Link https://blog.last9.io/sre-tooling-clever-hans-fallacy/

Clever Hans Fallacy of Monitoring Tools

Posted by nishantmodak on Wednesday July 29, 2020 @03:03AM
from the self-post dept.

All our monitoring tools enable us to observe what we want to look at. Known knowns.Thus, missing out on the unknown unknowns and being in an endless trial and error state. One at a time.https://blog.last9.io/sre-tooling-clever-hans-fallacy/

( Read More... | 1 comment | 6 points | on HN )
Re:Someone left my Gmail in debug mode on Tuesday December 01, 2015 @05:49AM (#10654534)
I can see this in my inbox too!
Re:Who Will Suffer Most from Climate Change? (Hint: Not You) on Friday September 04, 2015 @07:28AM (#10169763)
The rice farmers I met in Bihar, for instance, are now growing a new variety of flood-tolerant rice—nicknamed “scuba” rice—that can survive two weeks underwater.

Found this extremely interesting. Probably because, I was surprised by the thought that, this could be done

Re:Let's Talk About Beacons on Tuesday November 04, 2014 @02:42PM (#8558606)
Agree! With http://tah.io/get we have tried to make use of BLE not just as beacons but also for combining the power of your physical sensors with 'smartphone' sensors to have much more interesting use cases!
Re:Let's Talk About Beacons on Tuesday November 04, 2014 @02:26PM (#8558516)
(Disclaimer: Work at TAH) I totally agree to what you are saying. We have had a similar experience working with our devices and SDK.

http://tah.io/get

Re:Have I been pwned? Check if your email has been compromised in a da... on Thursday December 05, 2013 @01:02AM (#6852931)
You should also include dates alongside the breach. Just so that one knows, if they have taken corrective action after that
Re:Namecheap $0.98 com/net per year coupon on Monday May 27, 2013 @08:14AM (#5774359)
Interesting to see bitly stats on the link https://bitly.com/retweetdeals+
Re:Google’s $10 million 10,000 start-ups plan for India on Monday March 18, 2013 @09:11AM (#5393865)
$10 million . 10k startups.

$1000 in Google Adwords + Google Apps credits.

This is what they mean!

Re:Ian Ross, Who Led Bell Labs, Dies at 85 on Sunday March 17, 2013 @05:28AM (#5388664)
“If we had had the same progress in the aircraft industry, you and I could be flying between London and New York in 500,000-seat planes and the fare would be about 25 cents,” he once said.

What kind of progress/innovation does he refer to in this?

Older »