Shapoor Mistry supports RBI's push for Tata Sons public listing

Shapoor Mistry, whose Shapoorji Pallonji Group holds an 18.4% stake in Tata Sons, has voiced support for the Reserve Bank of India's directive regarding the conglomerate's listing. Mistry stated that the RBI's decision has brought 'full clarity' to the matter. He expressed willingness to work constructively with both Tata Sons and the Tata Trusts on the holding company's future direction. His comments come amid an ongoing power struggle within the Tata Group.

Developer builds gamified habit tracker with microservices, XP streaks, and resilience-first design

A software developer has built a habit-tracking application called Gamified Tracker that awards experience points using a formula combining activity duration, multipliers, and streak bonuses. The system is designed around a microservices architecture using Java 17 and Spring Boot 3.5, with separate services handling activity logging and gamification. A key design priority was resilience: habit logs must save successfully even if the XP or gamification layer fails, using an outbox pattern with RabbitMQ for asynchronous processing. The stack also includes Spring Cloud Gateway for routing and authentication, Netflix Eureka for service discovery, PostgreSQL for storage, and Prometheus with Grafana for observability. The developer acknowledges that a simpler monolith would suffice for current traffic, but chose the distributed setup deliberately to demonstrate and learn production-grade patterns.

MiRemoteBridge

Discussion | Link

Bhaker, Toor named India’s Asian Games flag-bearers

Kerala Expands Probe Into Messi Event; Ex-Sports Minister May Face Questioning

Kerala authorities have widened their investigation into alleged irregularities surrounding a planned Argentina national team match featuring Lionel Messi in Kochi in 2025. The probe centers on decisions made by the previous state government regarding the hosting of the high-profile event. A former sports minister is likely to be questioned as part of the expanded inquiry. The investigation aims to examine whether proper procedures were followed in organizing the match.

Ex Union minister Ravneet Singh Bittu's convoy attacked in Punjab

Some Anthropic Engineers Reportedly Treat AI Model Claude as a Deity

Reports from Silicon Valley suggest that certain engineers at AI company Anthropic have developed an almost reverential attitude toward their AI model, Claude. The rumor has sparked broader questions about how humans perceive and relate to increasingly powerful AI systems. Some technology leaders have contributed to this trend by describing advanced AI in god-like terms. Experts warn that treating AI as an unquestionable authority poses significant risks. The intense work culture at AI firms is believed to be blurring the boundaries between professional life and technology worship.

Humpback Whale Carries Stillborn Calf Off Australia in First-Ever Documented Case

Researchers off the Gold Coast of Australia observed a mother humpback whale displaying attentive behavior toward her stillborn calf on the ocean floor. The behavior persisted for several hours before scientists were able to document the event. This marks the first scientifically recorded instance of such conduct in humpback whales. The rare observation offers new insight into the depth of maternal bonds in these marine mammals. Scientists suggest the behavior may also indicate a form of grief among humpback whales.

Open-Source Researcher Claims Frontier Lab Replicated His Non-Autoregressive AI Work

A developer published two arXiv papers in March and September 2025 detailing non-autoregressive decision models trained with reinforcement learning, releasing open weights, datasets, and a PyPI package. In September 2026, well-funded startup TypeSafe AI launched a product called Jev, which the researcher says applies the same core concept without open weights, technical papers, or public training data. The researcher argues that generative LLMs are overkill for simple classification tasks like routing or jailbreak detection, where faster, lighter models suffice. In response, he built an open-source alternative called RL Agent on a bidirectional encoder, achieving 33–38ms inference latency compared to Jev's reported 150ms. The project is fully open-source and aims to demonstrate that non-autoregressive decision models are a practical, accessible alternative to large generative systems.

L&T shows 40 fabs & 1st silicon carbide chip for EVs

Two Indian Women from Andhra Pradesh Die in Australia Car Crash Involving Porsche

Two women from Andhra Pradesh, Neha Bommali (33) and her mother-in-law Rama Devi (67), have died following a road accident in Australia. Their vehicle collided with a Porsche, and all those involved were transported to a nearby hospital. Both women were declared dead on arrival by doctors at the hospital. The driver of the Porsche was also hospitalised following the crash.

Corporate Bitcoin Buying Slows to 5,900 BTC in Three Months Amid Weak Demand

Corporate treasuries purchased only 5,900 bitcoin over a three-month period, signaling a notable slowdown in institutional accumulation. The figures reflect a broader cooling in demand for bitcoin from corporate buyers. Other key demand indicators beyond treasury purchases have also shown weakness during this period. The combined data points suggest that enthusiasm for bitcoin among corporate entities has dimmed considerably in recent months.

Grafana as Code: Surviving Destroy-Recreate With Provisioning Files

Liquid syntax error: Unknown tag 'endraw'

Bengaluru GST Officer Caught on Camera Accepting Rs 8 Lakh Bribe

A GST officer in Bengaluru was caught on camera allegedly accepting a bribe of Rs 8 lakh. The arrest followed a complaint lodged with the Bengaluru City Lokayukta. Acting on the complaint, Lokayukta officials set up a trap to apprehend the accused. The officer, identified as Singh, was allegedly caught receiving the money through a mediator.

AEXGrid

Discussion | Link

Developer Builds Real-Time Token Scanner Pipeline for Robinhood Chain

A developer is building a real-time token scanner for Robinhood Chain, initially focused on tracking token activity around Pons. The pipeline moves through distinct stages — event listening, decoding, normalization, filtering, and API exposure — with each layer handling a specific responsibility. A key design principle separates the scanner's role of detecting activity from any downstream trading strategy, making the infrastructure reusable. The system is engineered from the start to handle real-world failure scenarios such as dropped connections, RPC provider outages, and duplicate event delivery. Idempotency and validation close to the ingestion layer are highlighted as critical to maintaining data reliability across the entire pipeline.

Developer argues AI tools sharpen skills rather than dull them

A debate exists among developers over whether relying on AI tools diminishes their technical abilities over time. Some practitioners worry that AI assistance causes their core skills to become rusty or underdeveloped. However, one developer argues the opposite, crediting AI with significantly improving their overall capabilities. They report resolving problems more quickly and discovering solutions at a faster pace than before. This experience, they say, has accelerated their learning across multiple areas of development.

Wombo

Discussion | Link

Why Running AI Agents in Production Requires a New Infrastructure Playbook

Platform engineers often assume AI agents can be deployed like standard containerized microservices, but this approach breaks down quickly in production environments. Unlike typical services, agent requests can run for minutes, carry state across multiple tool calls, and return HTTP 200 status codes even when their outputs are semantically wrong. Common Kubernetes defaults — such as liveness probes with 30-second thresholds and CPU-based autoscalers — are poorly suited to agents that may legitimately pause for 40–90 seconds during a reasoning step. Proper agent infrastructure requires separating process-liveness checks from forward-progress monitoring, setting timeout budgets per tool-call chain rather than per request, and treating session isolation as an architectural concern. The core argument is that agent infrastructure demands its own discipline, distinct from the microservice patterns that platform teams have relied on for the past decade.

Developer Builds Local-Only Secret Manager to Stop AI Coding Assistants Leaking API Keys

AI coding assistants read entire repositories to function, which means API keys stored in .env files can inadvertently end up in chat transcripts, logs, or model context windows. A developer identified this growing security gap and built an open-source tool called Concealer to address it. Concealer is a single Python script that encrypts secrets locally on the user's machine, allowing AI agents to use credentials without ever seeing their actual values. Unlike cloud vaults or password managers, it requires no account, no network connection, and is portable across machines with a single password. The project is publicly available on GitHub and targets solo developers and teams who rely heavily on AI-assisted coding workflows.