Aman Agarwal

Cyber security using AI @transilienceai | Founder of @openlit_io

[Aman Agarwal @_typeofnull]

@claudeai I think claude code hallucinated 😂

[Aman Agarwal @_typeofnull]

@claudeai I saw this error suddenly while viewing https://t.co/hd29WUWR3V . What happened ?

[Aman Agarwal @_typeofnull]

I wrote about a hard truth in RAG:
👉 Retrieval doesn’t eliminate hallucinations.

Even with good context, models can still give wrong answers.
In this blog:
* Where hallucinations come from
* Why retrieval isn’t enough (Dont trust the model!)

🔗 https://t.co/X5dxDIAenc

[Aman Agarwal @_typeofnull]

Been working with LLM systems lately and one thing stands out:

We track latency, tokens, cost… but barely look at the GPU.

which is where all the work happens.

Most debugging pain I’ve hit comes from this gap.

I wrote this blog to connect the dots:
https://t.co/QxiUXLbjFG

[OpenLIT 🔭 @openlit_io]

We wrote a quick guide on adding observability to LLM apps with OpenTelemetry.

Trace prompts, debug issues, and track latency/cost in minutes 👇

https://t.co/SYBI4VQylo

[OpenLIT 🔭 @openlit_io]

🚀 Python SDK v1.38.2 is out!

- Span name filtering
- VictoriaMetrics support
- Fix for eval LLM temperature
- Open WebUI pipeline improvements
- Dependency upgrades

Shoutout to new contributors 🎉

#AI #OpenSource #DevTools #openlit #llmops

[transilienceai @transilienceai]

Void Manticore didn't hack Stryker.
They logged into Stryker.

One phished token. Global Admin created at 3:14 AM. Every CA policy disabled. Bulk wipe- 214,839 devices, 79 countries, 28 minutes.

We rebuilt the entire kill chain as a 15-page manga. Page and PDF in the thread 🧵

Research sourced from @_CPResearch_ @Unit42_Intel @cloudsa

#VoidManticore #Handala #Stryker #MicrosoftIntune #EntraID #ThreatIntelligence #ZeroTrust #InfoSec

[Carly Richmond @CarlyLRichmond]

Come along to the DevOps Society meetup. There's talks on challenges running K8 clusters in prod, and myself on observing AI applications with @openlit_io OTel signals and @elastic.

https://t.co/T1IYw7xwKq

See you there! 👋

[Aman Agarwal @_typeofnull]

🚀 @openlit_io : TypeScript SDK (12+ integrations) + Go SDK (OpenAI + Anthropic) just launched.

Python, TS, Go. One line. Full observability.

Checkout https://t.co/NJA2j9znpW

#OpenLIT #Go #TypeScript #LLMObservability

[Aman Agarwal @_typeofnull]

Just dropped an episode on the AI Engineering Podcast with Tobias Macey
We covered what it takes to ship LLM apps to production — observability, cost tracking, vendor-neutral, and more @openlit_io
https://t.co/KplKAUr9lG
⭐ https://t.co/JPJK3n6dDT
#OpenTelemetry #AIEngineering

[Aman Agarwal @_typeofnull]

@Shashwat_12304 @claudeai

[SHASHWAT SHARMA @Shashwat_12304]

We were copy-pasting the same claude code skills across 5 projects like cavemen. update one, forget the other four, ship a bug on friday at 5pm. classic.
so we did the only logical thing, asked claude to build a project manager for claude projects.
introducing cldpm. open source, works today.
Github: https://t.co/j9kWn7Cw64
PyPI: https://t.co/mcFdXWrXqq
npm: https://t.co/oGVFy3q841

#ClaudeCode #AI #OpenSource #DevTools #BuildInPublic

[Elastic Dev @elastic_devs]

Adding application instrumentation is important because it helps us diagnose issues and bottlenecks. AI agents are no different!

OpenLit can be configured within your agent to send #telemetry information to Elastic via the following initialization:

[Carly Richmond @CarlyLRichmond]

@elastic_devs 2025 was definitely an AI learning year for me. I enjoyed building my first #typescript agent leveraging #elasticsearch and context engineering, and observing another with @elastic and @openlit_io:

https://t.co/ygkh5mRgzM

https://t.co/88Xz6sbTeK

[Aman Agarwal @_typeofnull]

We’re live on the @digitalocean Marketplace! 🚀
Deploy @openlit_io in one click and start building your LLM-powered apps and no ops hassle, just innovation.
🔗 https://t.co/THHH1pn9we
#OpenLIT #DigitalOcean #AI #LLM #SaaS #DevTools

[OpenLIT 🔭 @openlit_io]

We’re live on @FazierHQ!

🎉 Our zero-code Agent Observability is now launching on Fazier, go check it out!!

👉 https://t.co/wF8AUGIa4J

#AI #observability #opensource #Agents

[AI Native Foundation @AINativeF]

OpenLIT's Zero-code LLM Observability

🏅 Product Hunt Data

Ranking: 2025W41-rank45
Upvote: 149

🚀 Product Overview

OpenLIT offers zero-code observability for LLM applications and AI agents, allowing teams to monitor performance across models, vector databases, and GPU resources without modifying code or redeploying services.

📋 More Details

Built for teams managing production LLM workflows, OpenLIT eliminates friction by supporting both Kubernetes-native (via Helm) and cross-platform CLI installation. It works with existing tools like OpenAI, Anthropic, and LangChain, with flexible instrumentation and no vendor lock-in.

📊 Evaluation

AI Native Application Modernization: 91/100
OpenLIT is deeply AI-native, built specifically for LLM-based pipelines. Its zero-code model supports tight integration with language agents and prompt engineering, enabling rapid iteration in AI-based services.

🔗 Website

https://t.co/I8aMUNlufC

@openlit_io

[OpenLIT 🔭 @openlit_io]

Hey Everyone, We’re officially live on Product Hunt!

📷Introducing Zero-Code Observability for LLMs and AI Agents.

If you find our product useful, a quick upvote would mean a lot and help us reach more people:
https://t.co/oekxqOYudo

[Aman Agarwal @_typeofnull]

Hey 👋 My team and I just launched a product that makes it easy to track how AI tools perform — no setup changes and is Free! Takes under 5 minutes to use and works with most AI systems. Would mean a lot if you could check it out and upvote us here: 👉 https://t.co/Nn6TTEtbuE

[OpenLIT 🔭 @openlit_io]

🚀 Running LLMs on your own GPU?

Monitor memory, temp & utilization with the first OpenTelemetry-based GPU monitoring for LLMs.

⚡ OpenLIT tracks GPU performance automatically — focus on your apps, not hardware.

👉 https://t.co/BTEpV3epGJ

#LLMObservability #OpenTelemetry