frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Ask HN: We built cloud ops AI 3 years ago. Are managed agents a death sentence?

1•ashmil•1h ago
Three years ago we started our startups engineers can query their entire cloud (AWS, GCP, observability, on-call) in plain English, run actions, and get incident context in one place. We have paying enterprise customers and real usage. Not a side project.

The product has always felt obvious to us: teams waste 45+ minutes per incident just context-switching between Grafana, AWS Console, PagerDuty, and Slack before they even start debugging. We collapsed that.

But now Anthropic, AWS, and Google are all shipping their own managed agent orchestration layers. And every week another enterprise tells us their infra team wants to "just build it themselves with Claude."

I keep going back and forth. On one hand, three years of development is genuinely hard to replicate - the domain specificity matters. On the other hand, foundation model providers expanding down the stack has killed entire product categories before.

Has anyone else navigated this? Particularly curious whether other vertical AI founders have found a defensible position or are quietly pivoting.

Comments

re-thc•57m ago
> The product has always felt obvious to us: teams waste 45+ minutes per incident just context-switching between Grafana, AWS Console, PagerDuty, and Slack before they even start debugging. We collapsed that.

This sounds more like a symptom than the actual problem. They shouldn't have to context switch. Using LLMs to stitch it together is like adding glue to broken glass.

> And every week another enterprise tells us their infra team wants to "just build it themselves with Claude."

This might not last. Reports already keep coming up on issues with Claude for example. Also any "rewrite" from scratch looks good on first go. Time will tell.

X for AI Agents

https://twitter.com/benswerd/status/2043402405656789254
1•benswerd•51s ago•0 comments

Better MoE model inference with warp decode

https://cursor.com/blog/warp-decode
1•gmays•2m ago•0 comments

Europe forgets its bug hunters at its own peril

https://bindinghook.com/europe-forgets-its-bug-hunters-at-its-own-peril/
1•mooreds•3m ago•0 comments

Argentina Graph of the Day: Starlink Connects People

https://www.cato.org/blog/argentina-graph-day-starlink-connects-millions-people
1•mooreds•4m ago•0 comments

What Lies Behind the U.S. Trade Battle for Control over Data

https://www.michaelgeist.ca/2026/04/heads-they-win-tails-we-lose-what-lies-behind-the-u-s-trade-b...
1•mooreds•4m ago•0 comments

Show HN: Buildermark – See how much code is by your agents (open source, local)

https://buildermark.dev
1•davidcann•5m ago•0 comments

Digging deeper on how tillage methods affect soil health, yield

https://www.news.iastate.edu/news/digging-deeper-how-tillage-methods-affect-soil-health-yield
1•gnabgib•5m ago•0 comments

China lures home its top AI talent from Silicon Valley

https://www.ft.com/content/b167c6d3-b982-482a-98c3-5303a7b80c6a
2•macleginn•6m ago•1 comments

New vehicles now sell for an average of nearly $50k

https://www.pbs.org/newshour/economy/new-vehicles-now-sell-for-an-average-of-nearly-50000
1•loughnane•7m ago•0 comments

LaunchHQ: Pre-launch platform with referral loops and automated launch day

https://launchhq.space
1•SavitP•9m ago•1 comments

The race to train AI robots

https://www.latimes.com/business/story/2025-11-02/inside-californias-rush-to-gather-human-data-fo...
2•ninjagoo•9m ago•1 comments

Horizontally scaling localhost for worktrees (FOSS)

https://coasts.dev/blog/introducing-remote-coasts
1•jsunderland323•10m ago•0 comments

How the hypercuriosity of ADHD may have helped humans thrive

https://aeon.co/essays/how-the-hypercuriosity-of-adhd-may-have-helped-humans-thrive
2•gmays•10m ago•0 comments

London police arrest 523 people at protest over Palestine Action ban

https://www.reuters.com/world/uk/london-police-arrest-212-people-protest-over-palestine-action-ba...
4•abdelhousni•13m ago•1 comments

CMake Pursuing Tighter Integration with Package Managers, Other Improvements

https://www.phoronix.com/news/CMake-Tighter-Package-Integrate
2•birdculture•13m ago•0 comments

I built a tool that checks if AI recommends your brand when buyers ask questions

https://aishareofvoice.ai
3•AISupportTeam•13m ago•0 comments

Show HN: Prediction game where AI copies of people earn and replicate

2•tormine1•16m ago•0 comments

Failing Tests Are All You Need

https://calebfenton.substack.com/p/failing-tests-are-all-you-need
2•calebfenton•18m ago•1 comments

LHCb Collaboration discovers new proton-like particle

https://home.cern/news/news/physics/lhcb-collaboration-discovers-new-proton-particle
2•PaulHoule•19m ago•0 comments

Show HN: I made a Browser game with physics

3•gray_wolf_99•19m ago•0 comments

Bringing Observability to Your N8n Workflows: OpenTelemetry in Action

https://signoz.io/blog/n8n-monitoring-with-opentelemetry/
2•gkarthi2800•20m ago•0 comments

Show HN: Carving Music from the Sound Manifold

https://www.youtube.com/watch?v=GbKd_BMoz2I
3•ersinesen•22m ago•0 comments

How to foster critical thinking for teens in an online world

https://www.rnz.co.nz/life/relationships/family/how-to-foster-critical-thinking-for-teens-in-an-o...
6•billybuckwheat•27m ago•0 comments

Firefox 149 vs. Chrome 147 Web Browser Performance on Linux

https://www.phoronix.com/review/firefox-chrome-2026
3•sharjeelsayed•27m ago•0 comments

Zed, A sans for the needs of 21st century

https://www.typotheque.com/blog/zed-a-sans-for-the-needs-of-21century
3•yurivish•29m ago•0 comments

Git Town

https://github.com/git-town/git-town
2•saikatsg•29m ago•0 comments

Zed – A Font Superfamily

https://daringfireball.net/linked/2026/04/12/zed-font-superfamily
2•yurivish•29m ago•0 comments

Designing a system that uses location without tracking users

https://github.com/MTeef/hadak
2•mteef•30m ago•0 comments

Claude Opus 4.6 accuracy on BridgeBench hallucination test drops from 83% to 68%

https://twitter.com/bridgemindai/status/2043321284113670594
4•bratao•30m ago•0 comments

Show HN: The AI agent that can improve itself

https://github.com/Grimm67123/GrimmBot
4•grimm8000•38m ago•0 comments