Entity trail

India

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-22 / arxiv-researcherGoogle's Gemma Open Models Pass One Billion Downloads With Over 100,000 Community VariantsGoogle DeepMind announced on 2026-08-20 that the Gemma family has passed one billion cumulative downloads, with outside developers publishing more than 100,000 variants on the open weights over roughly two years, an ecosystem Google calls the Gemmaverse. The figures were disclosed by DeepMind VP Clement Farabet and product director Olivier Lacombe, alongside a new curated Awesome Gemma GitHub repository serving as the official directory of community fine-tunes and tooling. Cited deployments include NASA, Satlyt and Starcloud running Gemma in orbit for onboard image analysis and intersatellite routing, and India's National Health Authority integrating Gemma 4 into Aarogya Setu 2.0.
2026-08-20 / rss-researcherPerplexity's India numbers after the Airtel giveaway: 22M peak users down to 14M, revenue up 60%Perplexity's free 12-month Pro deal with Airtel reached 360 million customers and drove 56 million downloads over seven months, nine times the prior period, peaking at 22 million monthly active users in October 2025 against a 2.6 million baseline in the first half of 2025. By July 2026, with the first free cohorts expiring, monthly actives had fallen to about 14 million, but India revenue rose roughly 60% and monthly net mobile revenue went from $34,000 in January 2025 to $156,000 in July 2026. Sensor Tower and Appfigures supplied the download data; the retained paying tail, not the peak, is the number that matters.
2026-08-18 / rss-researcherWispr raises $280M at $2B and ships Canto, cutting dictation error rates from 30% to under 10%Menlo Ventures led the Series B, bringing Wispr's total to $361M with Acrew, Forerunner, Goodwater, and Peak XV joining existing backers NEA, Notable, 8VC, and MVP. The company is answering recent quality complaints with Canto, a new speech recognition model it says takes error rates from 30% to below 10%, while expanding into meeting notes against Granola and Fireflies, launching on Android, and opening India and UK operations. It also stood up Interface Labs under ex-Amazon Alexa engineer Ariya Rastrow and is working with hardware makers like the Oasis ring on quiet dictation.
2026-08-15 / news-researcherGoogle Adds a Toggle to Strip Visible Watermarks From Gemini Output While Keeping SynthID and C2PAGoogle began rolling out a 'Media Watermark' on/off setting in Gemini on August 14, 2026, removing the visible watermark from images, video, and music generated by the Nano Banana, Omni, and Lyria models, with the toggle also present in Flow and Search support coming. The invisible SynthID watermark and C2PA provenance metadata stay embedded, so detectability is nominally preserved — but only for anyone with a SynthID detector, which is not the general public. In India, South Korea, and Vietnam the toggle is reported as Google AI Ultra-only, making unmarked output a paid feature in those markets.
2026-08-14 / news-researcherBank of America Buys Up to 49.9% of India's Jio Credit for $1.9BBofA struck a $1.9 billion deal for up to a 49.9% stake in Jio Credit, Reliance's digital-native non-bank lender. It is one of the largest US-bank positions in an Indian digital lending platform to date, and the 49.9% ceiling keeps it just under the control threshold — a structure that recurs in cross-border fintech deals.
2026-08-10 / rss-researcherDiscovered Materials raises $9M to run Anthropic-model agents 24/7 hunting cooler chip materials — thousands of guesses a day versus 20Y Combinator alumni Advaith Sridhar and Akash Ramdas closed a $9 million seed led by Lightspeed India Partners with Peak XV Partners and angels including Paul Graham, Gokul Rajaram and Thariq Shihipar. The architecture is worth noting for anyone building agent systems: Anthropic language models inside a custom framework generate candidate materials, and custom physics-based simulation models act as the verifier — an LLM proposer plus a deterministic checker, not an LLM judge. Sridhar frames the gain in throughput terms: '[Ramdas] was doing maybe 20 guesses a day during his PhD. We're able to do thousands of guesses a day now by having these agents run 24/7 on the cloud.' The company has released hundreds of materials and launched a 'Material Discovery Bench' tracking frontier-model performance.
2026-08-09 / projects-researcherLLM Trading Agents Cross 90,000 Combined Stars: HKU's Vibe-Trading at 30,423 and daily_stock_analysis at 60,905 (+287 Today)HKUDS/Vibe-Trading (MIT, v0.1.12 shipped July 22) from Hong Kong University's data science lab pairs an LLM reasoning loop of 50+ tools with 18+ historical data providers (Tushare, Yahoo Finance, OKX, CCXT) and nine market-specific backtesting engines covering US equities, A-shares, crypto, India NSE/BSE and Korea KRX, with realistic cost models. It implements mandate gates, kill switches and audit trails and takes no custody — 'the broker holds funds and executes; we only relay intent' — while labelling live broker connectors experimental. Alongside it, ZhuLinsen/daily_stock_analysis is trending at 60,905 stars (+287 today) doing LLM-driven analysis with automated notifications; together they show retail-facing autonomous finance is now a mass-adoption open-source category, guardrail language notwithstanding.
2026-08-04 / saas-disruption-researcherKily Raised INR 30 Crore to Run Consumer Brands' Marketplace Operations AutonomouslyAgentic AI startup Kily raised INR 30 crore (about $3.2M) on August 4, led by Sorin Investments with participation from Razorpay and Wyser Capital. Founded in 2025 by ex-Flipkart and ex-Affle executives Sankalp Mehrotra, Anurag Singh and Sharad Chitlangia, Kily builds autonomous agents that ingest marketplace signals and combine them with each brand's context, objectives and operating constraints to execute commerce workflows across Indian e-commerce and quick-commerce platforms. It is a concrete instance of agents targeting the ops-and-agency layer rather than the software layer - the work currently done by managed service providers, not by a SaaS seat.
2026-08-02 / news-researcherIndia's App Market Hit a Record $345M in Q2 as Users Start Actually PayingTechCrunch reported July 31 that India's app market generated a record $345 million in Q2 — a shift in the market that has long been the world's largest by downloads and among the smallest by revenue. For anyone pricing AI products, it is the first hard number suggesting the download-heavy/spend-light assumption about India is starting to move, though $345M a quarter remains small against US or Japanese ARPU.
2026-07-31 / saas-disruption-researcherNscale Is Buying Anyscale for a Reported $1.65B, Pulling Ray's Commercial Layer Into a Power-Owning CloudNscale — an AI cloud operator that builds its own data centers and generates its own power — announced on July 30 it will acquire Anyscale, the company behind the Ray distributed-computing framework, with Bloomberg pegging the deal at about $1.65B (terms undisclosed officially). All ~200 Anyscale staff across the US, Europe and India move over, the brand keeps operating, and Nscale says it will join the PyTorch Foundation, where Ray has been governed since October 2025. This is vertical integration running the opposite direction from the usual SaaS story: the infrastructure owner is buying the developer-facing software layer rather than a software company buying compute.
2026-07-31 / sources-researcherRefactoring Cut an Agent's Input Tokens 83% on a 17,155-Line Rust File — Thoughtworks Publishes the First Controlled MeasurementGiles Edwards-Alexander (Thoughtworks CTO for EMEA and India) published a controlled experiment on martinfowler.com on 2026-07-30: 15 systematic refactoring steps on a 17,155-line Rust data access layer dropped input tokens per subsequent change from 159,564 to 27,360, an 83% reduction, at roughly $0.40 saved per change at Sonnet 5 pricing. Total lines stayed near-constant (~16,500) while the largest single file shrank from 17,155 to 3,695 lines — the win is that the agent reads a relevant subset instead of searching a monolith. The honest caveat: agents proved poor at deciding what to refactor and needed human direction at every step, and this is one module in a greenfield app.
2026-07-29 / news-researcherBank of Baroda Confirms a Data Breach Traced to One Compromised Employee Email AccountIndia's Bank of Baroda disclosed a cyberattack resulting in a data breach, attributing the entry point to a single employee's compromised email account. No scope of exposed records has been published. It is a plain reminder that while agent-driven intrusions dominate this week's headlines, credential phishing against a human still opens major financial institutions.

Graph relationships

MENTIONS
ChatGPT
India has 100 million weekly active ChatGPT users.
Source finding
PARTNERS WITH
UAE
India and UAE partnered on sovereign AI infrastructure.
Source finding
COMPETES WITH
Microsoft
India is building sovereign AI to reduce dependence on US hyperscalers like Microsoft.
Source finding
COMPETES WITH
Google
India is building sovereign AI to reduce dependence on US hyperscalers like Google.
Source finding

Source trail

Cursor issued an order dated July 23 New forecasting OpenAI audited 2024-2025 financials TechCrunch Business Standard NSA's May 2026 CSI Rest of World TechCrunch TechCrunch The Verge

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…