Abhishek Asthana
iabhishek.bsky.social
Abhishek Asthana
@iabhishek.bsky.social
Learning about things (mostly tech!) and sharing those learnings. AI enthusiast but not AI nerd!
Last night's Blood Moon. From India.

#BloodMoon #LunarEclipse
September 8, 2025 at 2:16 AM
Reposted by Abhishek Asthana
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.

By @iabhishek.bsky.social
AI Agents Can Talk, But Can We Trust Them?
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
bit.ly
September 1, 2025 at 3:30 PM
Reposted by Abhishek Asthana
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.

By @iabhishek.bsky.social
AI Agents Can Talk, But Can We Trust Them?
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
bit.ly
August 30, 2025 at 2:00 PM
Hello AI Bluesky. I wrote about the challenges facing agentic communication, especially across organisation boundaries.

What challenges do you see? Or how have you solved them?

#AI #AgenticAI
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
AI Agents Can Talk, But Can We Trust Them?
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
bit.ly
August 30, 2025 at 5:52 AM
Reposted by Abhishek Asthana
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
AI Agents Can Talk, But Can We Trust Them?
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
bit.ly
August 28, 2025 at 11:00 AM
My post about challenges facing agentic communication across organisation boundaries.

How do we know which agent did what? Is an agent really as capable as it claims? And how to handle malicious agents?

Do give a read. #AI #AgenticAI
thenewstack.io/ai-agents-ca...
AI Agents Can Talk, But Can We Trust Them?
AI agents can now communicate autonomously, but can we trust them? This article explores key challenges like audit trails, agent authenticity, and security risks.
thenewstack.io
August 29, 2025 at 3:20 PM
Oh, to be a wild mongoose lying flat on a log and just chilling!

www.instagram.com/share/p/BAUE...

#Wildlife #WildAnimals #IndianWildlife #Conservation
August 23, 2025 at 4:01 PM
The eyes have it!

Leopard at Tadoba tiger reserve, India.

#BigCat #Wildlife #Photography #Leopard
July 15, 2025 at 12:46 PM
Tiger with its kill at Tadoba tiger reserve, India

#Wildlife #Conservation
July 11, 2025 at 2:39 PM
I wrote a post on Agentic AI: Multi-agent architectures medium.com/@abhi.asthan...

#Agents #AgenticAI
Agentic AI: Multi-agent architectures
In this post on Agentic RAG we discussed how AI agents overcome most of the shortcomings of naive RAG. But as they say, what is better than…
medium.com
April 2, 2025 at 2:30 AM
I wrote about how AI Agents do RAG. What makes Agentic AI so powerful and what complexities it hides from users? Also included is code to create a simple Agentic RAG solution using CrewAI. medium.com/@abhi.asthan... #AI #AgenticAI #Agents
Agentic RAG with CrewAI
We have talked about developing a simple RAG application and various stages of RAG where we discussed what makes an advanced RAG system…
medium.com
March 15, 2025 at 5:05 AM
Reposted by Abhishek Asthana
Vibe-coded a quick tool to help me track how close the pelicans are to getting back home to Half Moon Bay - they've made it as far as Santa Cruz tools.simonwillison.net/species-obse...
March 15, 2025 at 3:49 AM
Reposted by Abhishek Asthana
Planet Definitions xkcd.com/3063
March 14, 2025 at 4:07 PM
I wrote about how AI Agents do RAG. What makes Agentic AI so powerful and what complexities it hides from users? Also included is code to create a simple Agentic RAG solution using CrewAI. medium.com/@abhi.asthan... #AI #AgenticAI #Agents
Agentic RAG with CrewAI
We have talked about developing a simple RAG application and various stages of RAG where we discussed what makes an advanced RAG system…
medium.com
March 14, 2025 at 4:54 PM
Reposted by Abhishek Asthana
Gemini has a new embedding model out called gemini-embedding-exp-03-07 - I've added support for that in llm-gemini 0.14, notes here simonwillison.net/2025/Mar/7/g...
State-of-the-art text embedding via the Gemini API
Gemini just released their new text embedding model, with the snappy name `gemini-embedding-exp-03-07`. It supports 8,000 input tokens - up from 3,000 - and outputs vectors that are a lot …
simonwillison.net
March 7, 2025 at 11:23 PM
I just published "Measuring vector similarity using Raw Cosine Similarity" medium.com/p/measuring-... #AI #LLMs #Medium #VectorAlgebra
Measuring vector similarity using Raw Cosine Similarity
One of the most basic problems in RAG involves comparing vectors and finding the most similar ones.
medium.com
February 26, 2025 at 3:33 PM
I just published this article.

When RAG is no more Naïve: Techniques used by Advanced RAG to improve results link.medium.com/UpiuRNFdPQb

#RAG #AI #LLM
link.medium.com
February 9, 2025 at 2:36 AM
Reading Andy Weir's "Artemis", I got enamored by Moon once again!
#Astrophotography
#Artemis
#Moon
January 1, 2025 at 1:56 AM
Hello Blue Sky.

Enjoy this photo of mamma Leopard and her cub.

#MasaiMara #Leopards #WildLife #WildlifePhotography
December 9, 2024 at 1:47 PM
Ever since GPT landed, search has changed just so much. Why not carry the experience to everything that can be searched? In this first part of two-part post I write about what is RAG and how it brings the goodness of LLM to everyone and everything.

lnkd.in/gese7Rx7

#LLM #GenAI #RAG #Python #Llama
LinkedIn
This link will take you to a page that’s not on LinkedIn
lnkd.in
December 8, 2024 at 2:28 PM
Reposted by Abhishek Asthana
Australia sensibly using their worst bowlers now, knowing that the India batters will be terrified of ruining their IPL auction prospects by getting out to one of them. #AUSvIND 🏏
November 24, 2024 at 3:57 AM
Hello Blue Sky.

So y'all love leopards?

This female has just killed a monkey and decided to have its meal in peace but was annoyed by all the chattering!

#Wildlife #WildlifePhotography
#India
November 22, 2024 at 1:56 AM