A 15-year-old dream has come true today. I started a PhD with the dream of creating a system that chants any Sanskrit shloka perfectly.
And here I am opening sourcing ๐๐๐ ๐๐ก๐๐ง๐ฎ - ๐ ๐ฏแน๐ญ๐ญ๐ (๐ฆ๐๐ญ๐๐ซ) ๐๐ฐ๐๐ซ๐ ล๐ฅ๐จ๐ค๐-๐ญ๐จ-๐๐ก๐๐ง๐ญ ๐ญ๐๐ฑ๐ญ-๐ญ๐จ-๐ฌ๐ฉ๐๐๐๐ก (TTS) ๐ฌ๐ฒ๐ฌ๐ญ๐๐ฆ ๐๐จ๐ซ ๐๐๐ง๐ฌ๐ค๐ซ๐ข๐ญ. This is the world's first vrutta-aware, open-source TTS for Sanskrit Chanting.
Apapang, Alang and Ahu, the tiny Amur Falcon trio, here are the faces behind the flight and wings behind the wonder ! You will see that while Ahu continues to move around in northern Somalia, both Apapang and Alang have moved on into Tanazania and Kenya respectively. Ahu has stopped at Somalia which is not surprising as they generally do after a long flight. Falcons start to feed again to put on fat reserves that they exhausted during their over sea migration. As told to @supriyasahuias by @sureshwii@wii_india #AmurFalconMigration #Amurfalcons
For those struk by the magic of Amurs ! Apapang makes it across the Arabian Sea with ease and enters the Horn of Africa, specifically Somalia. Apapang, being an adult male, has definitely done this oceanic crossing multiple times before. Apapang has now done nearly 5400 km, and it has taken him 5 days and 15 hours.
as told by @sureshwii
And the epic journey begins again in all its glory. Three new travellers, Apapang (adult male) orange track, Alang (young female) Yellow track, and Ahu (adult female) Red Track, were satellite-tagged on 11th November 2025 as part of the Manipur Amur Falcon Tracking Project (Phase 2) by @wii_india . In just days, Apapang has stunned trackers with an extraordinary non-stop flight, already cutting across central India and now skimming the Arabian Sea, poised for a 3,000 km oceanic crossing to Somalia,one of the most demanding journeys undertaken by any raptor on the planet. From the forests of Manipur to the vast African landscapes that await them, these tiny birds barely 150 grams continue to remind us of the sheer wonder of migration, and why Indiaโs protection of stopover sites has become a global conservation story. What a wonder ! Credits @sureshwii
#AmurFalcons #BirdMigration
India's nightlife is getting a spiritual remix with the rise of "bhajan clubbing,"
Young Hindus are swapping whisky shots for chai and techno beats for tabla drops, blending faith with fun. And yes, it has absolutely everything to do with religion.!
This is how ๐๐ ๐๐ด๐ฒ๐ป๐ ๐ ๐ฒ๐บ๐ผ๐ฟ๐ works.
In general, the memory for an agent is something that we provide via context in the prompt passed to LLM that helps the agent to better plan and react given past interactions or data not immediately available.
It is useful to group the memory into four types:
๐ญ. Episodic - This type of memory contains past interactions and actions performed by the agent. After an action is taken, the application controlling the agent would store the action in some kind of persistent storage so that it can be retrieved later if needed. A good example would be using a vector Database to store semantic meaning of the interactions.
๐ฎ. Semantic - Any external information that is available to the agent and any knowledge the agent should have about itself. You can think of this as a context similar to one used in RAG applications. It can be internal knowledge only available to the agent or a grounding context to isolate part of the internet scale data for more accurate answers.
๐ฏ. Procedural - This is systemic information like the structure of the System Prompt, available tools, guardrails etc. It will usually be stored in Git, Prompt and Tool Registries.
๐ฐ. Occasionally, the agent application would pull information from long-term memory and store it locally if it is needed for the task at hand.
๐ฑ. All of the information pulled together from the long-term or stored in local memory is called short-term or working memory. Compiling all of it into a prompt will produce the prompt to be passed to the LLM and it will provide further actions to be taken by the system.
We usually label 1. - 3. as Long-Term memory and 5. as Short-Term memory.
A visual explanation of potential implementation details ๐
And that is it! The rest is all about how you architect the topology of your Agentic Systems.
What do you think about memory in AI Agents?
#GenAI #AI #MachineLearning
McKinsey released their 76-page long report on generative AI.
They say AI will automate 30% of work hours by 2030.
Here are 5 most important things you need to know:
#Namaste BHU! ๐
Major boost to digital connectivity within the university, strengthening internal communication, and making essential information easily accessible to the university fraternity. Download #NamasteBHU app through @AppStore or @GooglePlay.
#BanarasHinduUniversity
The late Inder Bhan Madan's cartoons, on display at FICA, Lado Sarai, are a fascinating view at Indian modernity. More importantly, he represents what it means to do art for its own sake in an age of self-promotion
https://t.co/PHqNcTkZV6