prateekdalal

19 ms·

1.
by prateekdalal 7mo ago
Hi HN, Over the last 3 days, I shipped a major update to Kakveda (OSS). The goal: unify infrastructure monitoring, observability, Kubernetes visibility, profiling, DB monitoring, and RUM into a single open source stack. What’s included in v
2.
Kakveda open source-level AI, Infra observability agent(kakveda.com)
1 points by prateekdalal 7mo ago|1 comments
3.
by prateekdalal 7mo ago
Over the last 48 hours we shipped a major update to our open-source observability + infra + AI runtime stack. Big changes: Renamed kakveda-aankh → kakveda-netra Built as a real host-level agent (CLI + systemd support) Background mode (--sta
4.
by prateekdalal 7mo ago
We’ve released v1.0.2 and v1.0.3 of Kakveda (open source AI governance runtime). This release shifts integration from manual middleware to an SDK-first model. Key change: Instead of calling /warn and /publish manually, agents now
5.
Kakveda v1.0.3: SDK-first governance layer for AI agents(github.com)
1 points by prateekdalal 7mo ago|1 comments
6.
Failure Intelligence for AI Systems(github.com)
1 points by prateekdalal 7mo ago|1 comments
7.
by prateekdalal 7mo ago
Over the past year, I’ve noticed something interesting in production AI systems: Failures don’t just happen — they repeat. Slightly different prompts. Different agents. Same structural breakdown. Most tooling today focuses on: Prompt qualit
8.
by prateekdalal 7mo ago
Thanks for sharing — that article points at the same core tension. Determinism isn’t about rejecting probabilistic systems, it’s about deciding where uncertainty is allowed to live. What keeps breaking in practice is when probabilistic reas
9.
by prateekdalal 7mo ago
I mostly agree — using LLMs to help author deterministic code is a good fit. The distinction I’m trying to draw is that once that code exists, the determinism has to live outside the model. LLMs can assist in creating the rules, but they sh
10.
by prateekdalal 7mo ago
I doubt ISO-9000 gets “replaced” so much as interpreted more strictly in the presence of LLMs. ISO-9000 isn’t about how work is done — it’s about whether processes are defined, repeatable, auditable, and improvable. From that lens, LLMs act
11.
by prateekdalal 7mo ago
I don’t disagree with the underlying concern. In practice, “probabilistic” often does translate to unreliable when you put these systems in environments that expect reproducibility and accountability. Where I think the framing matters is in
12.
by prateekdalal 7mo ago
That sounds very aligned. I like the way you phrased it - deterministic policy that agents can not bypass is exactly the right boundary, especially once you assume prompt injection and misalignment are not edge cases but normal operating co
13.
by prateekdalal 7mo ago
This resonates a lot, and I think your example actually captures the core failure mode really well. What your PM asked for isn’t an “agentic pipeline” problem - it’s an organizational knowledge and accountability problem. LLMs are being use
14.
LLMs are powerful, but enterprises are deterministic by nature
7 points by prateekdalal 7mo ago|15 comments
15.
Show HN: Kakveda – Failure intelligence and pre-flight warnings for LLM systems(github.com)
2 points by prateekdalal 8mo ago|0 comments