blog@Yantrajaal

রহস্যের আড়ালে প্রযুক্তি, আর প্রযুক্তির আড়ালে মানুষ গোয়েন্দা গল্প বললেই আমাদের চোখের সামনে ভেসে ওঠে খুনের জায়গায় পড়ে থাকা কোনও সূত্র, সন্দেহভাজন কয়েকজন মানুষ, আর এক তুখোড় গোয়েন্দা, যিনি শেষ পর্যন্ত সব রহস্যের জট খুলে দেন। কিন্তু অপরাধের দুনিয়াটা এখন অনেক বদলে গেছে। খুন করতে সবসময় বন্দুক লাগে না, কাউকে ফাঁসাতে ঘটনাস্থলে উপস্থিত থাকারও দরকার পড়ে না। একটা স্মার্টফোন, কম্পিউটার নেটওয়ার্ক, কিংবা কোনও নিরীহ দেখতে ইলেকট্রনিক যন্ত্রও হয়ে উঠতে পারে অপরাধের হাতিয়ার। এই বদলে যাওয়া পৃথিবীকে নিয়েই পাঁচটি রহস্যকাহিনির সংকলন ‘কালের ফাটল’। আর এই পাঁচ গল্পের কেন্দ্রে রয়েছে তিনজন একেবারেই অন্যরকম তদন্তকারী। পারলিকা  মুখার্জি, ওরফে পারো। চল্লিশ পেরোনো, স্বাধীনচেতা, বুদ্ধিমতী এক কলকাতার মহিলা। পেশায় ফরেনসিক অ্যাকাউন্ট্যান্ট। হিসেবের গরমিল ধরতে গিয়ে মানুষের মনের গরমিলও বেশ ভালোই বুঝে ফেলেন। জীবনের নানা ঝড়ঝাপটা সামলে নিজের মতো করে বাঁচতে শিখেছেন। ভালো খাবার, ভালো পানীয়, আড্ডা—কোনও কিছুতেই তাঁর আপত্তি নেই। তবে কোনও রহস্য মাথায় ঢুকে গেলে তার শেষ না দেখে ছাড়েন না। শিখা মাণ্ডি। বয়সে ...

 

Consider a crisis in the Strait of Hormuz. Iran closes the waterway. The United States must decide whether to respond with military force, negotiate, or seek some less obvious accommodation. Each decision could trigger a different chain of reactions, leading to escalation, stalemate or de-escalation. Can we build a model that systematically explores these alternative futures without pretending to know which one will actually occur? But to begin with, what on earth is a model? Models are proxies that help understand the behaviour of physical and social systems. A toy car helps us somewhat understand how real cars move, while Newton's laws explain gravity. Models evolve: a radio-controlled car is more sophisticated than a wooden toy, just as Einstein's model of gravity improves upon Newton's. Complex situations that are dependent on the outcome of probabilistic or chance events are often described by simulation models based on the Monte Carlo method. In these simulations, occ...

  Two presentations exploring the emergence of artificial intelligence and its implications for individuals, organizations and society. The first, Concepts & Contours, traces the evolution of AI from conventional machine learning to generative AI, and examines a fundamental question: as machines acquire capabilities once considered uniquely human, how will the relationship between humans and machines evolve? The second, Transformers & Transformations, moves from technology to its practical implications for business. It explores how AI can go beyond automating existing processes to enable entirely new ways of working, making decisions and organizing businesses. It also introduces the idea of a Centaur Culture, where human judgement and machine intelligence complement each other. The underlying theme of both presentations is simple: AI expands what is possible. Human imagination determines what is worth pursuing.

Parashar21 reached a milestone today, with a version that has reached the limits of its current architecture. The Rashi and Navamsha charts and the Vimshottari Dasha sequences are now being generated correctly — until, of course, an error is found. Gochar and Ashtakavarga are also being calculated, but for the moment I am leaving them outside the analysis and interpretation engine. The NotebookLM-based interpretation engine now works with two clean text corpora, from Braha and Harihar Majumdar. Unless NotebookLM upgrades its underlying LLM, there is little more that I can do at that end. The interpretations are reasonable, and any further improvement will probably come from experimenting with better prompts. At this stage, neither the computational engine nor the corpus can be meaningfully improved without changing the architecture itself. So this seems like a good place to stop. And so, on the day of the Autumn Equinox, I offer this labour of love at the feet of Goddess Saraswati, wit...

 Kanyā Saṅkrānti when the Sun enters Kanyā, and Bengal remembers Viśvakarmā, the divine architect. But this day belongs also to Manasā  serpent goddess of hidden knowledge, instinct and the powers beneath the visible world. So, for today, a Jugalmūrti: Viśvakarmā and Manasā — the maker of tools and the keeper of mysteries.

  Universities are very good at teaching the Known Knowns -- the accumulated knowledge of disciplines such as literature, mathematics, science, engineering and management. Increasingly, they also try to prepare students for the Known Unknowns: AI, climate change, sustainability, geopolitics and other challenges that we can already see approaching. But what about the Unknown Unknowns — the changes that we cannot anticipate because we do not yet know what they are? That question led me to TU3 -- The University of Unknown Unknowns , a thought experiment in curriculum design. Instead of trying to predict tomorrow's subjects, TU3 asks whether students can be equipped with capabilities that remain useful whatever tomorrow brings: to THINK critically, COMPREHEND deeply, and CONSTRUCT physically. These three abilities form the TU3 Tripod, a common core around which conventional undergraduate disciplines can be built. The university may be imaginary; the question behind it is quite real. Ex...

 Presentation delivered at the Education Summit hosted by Bengal Chamber of Commerce and Industry, at Taj Bengal on 3rd September 2026  

Using the Google Stack for LLM-Based Astrology Can an LLM-based AI provide astrological advice and guidance? Given the fact that LLM platforms like ChatGPT and Claude, along with their agentic AI extensions, have been demonstrating mind-bending performance in complex, cognitive tasks, the question is no more academic. If they can perform complex logical analysis of science and mathematics problems, astrology should be easy. Not quite. There are certain challenges.  Having looked at the issue of using computers to perform, or at least assist with, astrological predictions for over 20 years, the problem is fascinating. We know that the computational aspect -- calculating positions, etc., -- is completely deterministic and hence algorithmic. This can be done with conventional programming. However, the next part, the interpretation of horoscopes is extremely tricky because there are literally thousands and thousands of conditional situations -- programmatically referred to as IF-THEN-E...

----- For more such slide decks - see The Silicon Stack Rare Earth Elements

  Ever since OpenAI dropped ChatGPT into the world, things haven't been the same. Sure, AI as a concept goes back to Alan Turing -- the father of modern computing -- but ChatGPT forced it right into our daily conversations. Today, you can hardly talk about the future without bringing up AI: how incredible it is on one hand, and the darker fears on the other -- lost jobs, or even machines replacing human companionship. This obsession feels intimately familiar to me. It takes me back to the frenzy around computer science -- or IT -- at the turn of the century. Fueled by Y2K, India convinced itself that the direct path to corporate nirvana ran straight through companies like Infosys, TCS, and Wipro. Students rushed into IT from every branch of engineering, and soon from almost every other background too, leaving almost everything else behind. It didn't matter what you studied or if you even liked coding. What mattered was that doing basic software work for American companies was t...

The Śiva Kalpataru Hymn is an anonymous, late-classical or medieval devotional hymn rooted in regional Shaiva temple traditions. It is different from the Twelve-Syllable Bhujangaprayāta Śivāṣṭakam attributed to Ādi Śaṅkarācārya, that is composed in the Bhujangaprayāta meter (literally "the gliding motion of a snake"). The Śiva Kalpataru Hymn survived and spread not through royal courts or famous philosopher-saints, but through generations of devotees chanting it together as a living prayer to Lord Śiva, the Wish-Fulfilling Tree. This Hymn is particularly prominent in the devotional Shaiva traditions of Eastern and Coastal India -- most notably in Odisha, Bengal, and parts of Bihar and Chhattisgarh. Because it was preserved as an oral chant rather than a rigidly fixed scholarly treatise, the text exhibits classic features of a fluid folk-liturgical origin. This is apparent from Modular Stanzas: Over centuries, lines from other popular Shaiva hymns (such as descriptions of Trip...

 Ego, Shadow, and Medium 1 Vector Art / Silhouette 2. Stippling / Ink Pointillism 3. Charcoal / Chalk Drawing 4. Monochrome Watercolour 5. Pop Art / Half Tone 6. Vector Art / Silhouette 7. Etching / Engraving 8. Graphic Novel / Comic Book 9. Linocut / Block Print 10 Pencil / Graphite Sketch

Education has to meet students where they’re actually hanging out. Right now, Gen Z is getting a massive chunk of their information from a quick scroll through YouTube Shorts or Instagram Reels. Instead of fighting that reality, open universities should run with it, and the entire engine of this shift will be driven by a pure attention economy. Picture this: universities lay out their curriculum publicly, then open the doors for digital creators to turn those dense syllabi into sharp, engaging, bite-sized videos. But here is the kicker: creators get paid based entirely on raw audience retention, specifically, view counts and exact time spent on screen. This is the ultimate "Uberization" of higher ed, fueled by performance metrics. The dominant model shifts from standard licensing fees to a direct payout for keeping eyeballs glued to educational content. If a creator’s video hooks a student and keeps them watching till the final second, they get paid premium rates. If viewers...