frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

How are people debugging multi-agent AI workflows in production?

https://www.agentsentinelai.com/
1•skhatter•1h ago

Comments

skhatter•1h ago
I've been experimenting with AI agents and multi-step workflows recently and ran into a problem that reminded me a lot of early distributed systems.

Once agents start calling tools, APIs, and other agents in a chain, debugging failures becomes surprisingly hard. A single task can involve multiple steps—LLM calls, tool invocations, retries—and when something breaks it's often difficult to understand exactly what happened or where the failure originated.

In traditional distributed systems we eventually built things like tracing, circuit breakers, retry policies, SLOs, and other reliability primitives to operate systems safely in production.

I'm curious how people building agent systems today are handling this.

Some questions I'm particularly interested in: - How do you debug agent failures? - Do you have visibility into multi-agent workflows? - How do you replay or reproduce failures?

I've been exploring this problem space and built a small prototype to experiment with reliability tooling for agent systems. The link above shows the demo, but I'm mainly interested in learning how others are approaching this problem.

verdverm•1h ago
OTEL and LGTM, the same open source o11y stack I use for everything
skhatter•1h ago
Interesting — are you instrumenting the agent workflows themselves with OpenTelemetry spans?

I was wondering how well the standard o11y stack works once agents start running multi-step workflows (agent → tools → other agents → APIs). Tracing probably helps visualize the steps, but I'm curious how people handle operational things like retries, replaying failed workflows, or containing cascading failures across agents.

Those reliability aspects are what I've been exploring.

Show HN: The Common Infrastructure for Agentic Communication

https://cyrisai.dev/
1•krishnamzg•2m ago•0 comments

Adapting to the New AI Landscape in Engineering

https://docs.google.com/document/d/1-vxazqwALKGivwedpGET4j1ipgab09C6RbLMTZ5ve5U/edit?tab=t.wi7nr0...
1•alexjray•2m ago•0 comments

Fed to loosen capital requirements for big US banks

https://www.ft.com/content/a1c81f17-201f-4e3f-8e02-e0b304f1b6a1
2•petethomas•3m ago•1 comments

How HN: PDF Table Extractor – AI-powered tool to extract tables from PDFs to CSV

https://pdf-table-extractor-5wak.vercel.app
1•atdl•4m ago•1 comments

What's My ΔEOK JND?

https://www.keithcirkel.co.uk/whats-my-jnd/
1•donohoe•8m ago•0 comments

Agent Skills for Interview Preparation

https://github.com/jiito/interview-prep-skills
1•jiito•10m ago•0 comments

Probabilistic Execution Beyond Classical Systems

https://www.authorea.com/users/903147/articles/1391658-probabilistic-execution-beyond-classical-s...
1•huiwenhan•10m ago•1 comments

Judgment and creativity are all you need

https://lethain.com/judgment-is-all-you-need/
1•donutshop•11m ago•0 comments

Hackers reportedly stole nearly 1,000TB of data from Telus Digital

https://mobilesyrup.com/2026/03/12/hackers-steal-nearly-1000tb-of-data-from-telus-digital/
2•whynotmaybe•13m ago•0 comments

Management in the Age of AI

https://blog.staysaasy.com/p/management-in-the-age-of-ai
1•donutshop•13m ago•0 comments

To Sparsify or to Quantize: A Hardware Architecture View

https://www.sigarch.org/to-sparsify-or-to-quantize-a-hardware-architecture-view/
2•matt_d•29m ago•1 comments

Tennessee grandmother jailed after AI face recognition error links her to fraud

https://www.theguardian.com/us-news/2026/mar/12/tennessee-grandmother-ai-fraud
8•danso•31m ago•0 comments

Don't Vibe – Prove

https://ngrislain.github.io/projects/2026-3-12-dont-vibe--prove/
2•ngrislain•36m ago•0 comments

I built the Vy replacement that launches March 26th, the day Vy shuts down

https://inceptive-ai.com
2•alymaknojiya•36m ago•3 comments

Meta delays rollout of new AI model after performance concerns

https://www.nytimes.com/2026/03/12/technology/meta-avocado-ai-model-delayed.html
5•wibbily•39m ago•2 comments

Adobe's longtime CEO to exit role amid AI disruption, shares fall

https://www.reuters.com/sustainability/boards-policy-regulation/adobe-announces-ceo-transition-sh...
5•tartoran•43m ago•0 comments

One plan/spec to rule them all (at least replace lots of docs)

3•wek•45m ago•0 comments

In space, no one can hear you kernel panic

https://increment.com/software-architecture/in-space-no-one-can-hear-you-kernel-panic/
2•p0u4a•46m ago•1 comments

Rivian Introduces R2 Lineup, Sharing Full Trims and Pricing

https://rivian.com/newsroom/article/rivian-introduces-r2-lineup
3•freetime2•47m ago•1 comments

Show HN: Nix on Windows –- proof-of-concept demo

https://github.com/nix-windows/nix-windows-demo
7•Ericson2314•49m ago•0 comments

How a subtle CSP misconfiguration broke our admin panel and how we fixed it

https://syndicode.com/blog/csp-failure-rails/
2•lglazyeva•50m ago•0 comments

Auto-Browser – An MCP-native browser agent with human takeover

https://github.com/LvcidPsyche/auto-browser
1•Lvcid•52m ago•0 comments

Silicon Valley's New Obsession: Watching Bots Do Their Grunt Work

https://www.wsj.com/tech/ai/ai-bots-claude-openclaw-285ac816
2•bookofjoe•52m ago•1 comments

Silicon Valley Abuzz About Adding AI Compute to Engineer Compensation

https://www.businessinsider.com/ai-compute-compensation-software-engineers-greg-brockman-2026-3
1•healsdata•52m ago•0 comments

Show HN: Tarvos – Relay Architecture for infinitely building with coding agents

https://github.com/Photon48/tarvos/tree/main
1•Photon48•53m ago•0 comments

Enabling Efficient Sparse Computations Using Linear Algebra Aware Compilers

https://www.osti.gov/biblio/3013883
1•matt_d•53m ago•0 comments

Show HN: Parevo Core – Auth, tenant, permission in one Go library

https://github.com/parevo/core
1•parevo•55m ago•0 comments

Moving beyond RLM and ReAct based coding agents

https://randomlabs.ai/blog/slate
2•akira_067•57m ago•0 comments

Full Source Code of Sweden's E-Government Platform Leaked from Compromised CGI

https://darkwebinformer.com/full-source-code-of-swedens-e-government-platform-leaked-from-comprom...
2•reimertz•58m ago•0 comments

Musk's X to Alter Verification System in Europe

https://news.bloomberglaw.com/california-brief/musks-x-to-alter-verification-system-in-europe-com...
1•absqueued•1h ago•0 comments