Today’s lead · x.com
LongCat-2.0 open-sources a 1.6T-parameter MoE under MIT
Meituan LongCat says it released LongCat-2.0 model weights and inference code under an MIT license. The model is described as a 1.6T-parameter MoE with about 48B active parameters, 1M-token context, agentic coding integrations, and GPU/NPU deployment support.
Top signals
2 more
Amazon will close Mechanical Turk to new customers on July 30
TechCrunch reports AWS will stop accepting new Mechanical Turk customers on July 30, 2026, while existing customers can continue using the service. AWS says it will keep investing in security and availability but does not plan new features.
Report: Anthropic seeks at least 1.4GW of Australian data center capacity
A post citing leaked documents says Anthropic is looking to buy at least 1.4GW of Australian data center capacity, with the buildout potentially costing up to $15B.
Tools & repos
5 selectedZackriya-Solutions/meetily
Meetily is a self-hosted, local-first AI meeting assistant for macOS and Windows, built in Rust with live Parakeet/Whisper transcription, speaker diarization, and Ollama-based summarization. It is useful for teams that want meeting notes without sending audio to a cloud service.
TryCase
TryCase provides disposable test environments for AI coding agents. The core use case is giving agent-generated code a safe, isolated place to run and validate changes before they touch shared development or production systems.
T3MP3ST
T3MP3ST is a multi-agent offensive-security framework that connects to Claude Code, Codex, Hermes, or local models and runs authorized recon, exploit, and reporting workflows. Its README claims a 90.1% pass@1 result on XBOW's 104-challenge suite, with reproducible benchmark scripts.
Hermes MoA 2.0
Nous Research released Hermes Mixture of Agents 2.0, an open-source ensemble layer that queries models such as GPT, Claude, and DeepSeek in parallel and synthesizes a final answer. The useful signal is less another base model and more a configurable orchestration layer for reasoning, coding, and instruction-following tasks.
mails
mails is email infrastructure for agents, with a CLI/SDK for sending, receiving, searching inboxes, handling attachments, and extracting verification codes. It can use hosted @mails.dev mailboxes or a self-hosted Cloudflare Worker setup.
Blogs worth your time
1 reads
Hugging Face Kernels adds trusted publishers, signing, and a Hub repo type
Hugging Face explains the Kernels redesign: a first-class Hub repository type for native kernels, trusted publishers, Nix-based reproducible builds, Git SHA provenance, Sigstore signing, Torch Stable ABI support, TVM FFI, and agent-friendly kernel-builder workflows. Useful if you ship custom accelerator kernels and need distribution and security patterns before agents start generating kernel code.
Funding & acquisitions
1 moves
PRISM (OYO)
OYO parent Oravel Stays, now rebranded as PRISM, has filed an updated DRHP for a ₹6,650 Cr fresh-issue IPO with no offer for sale. The filing shows profitability and global diversification, but nearly 75% of proceeds are intended for debt repayment.
Bengaluru radar
4 eventsStartup World Cup 2026 Bengaluru
Pegasus Tech Ventures and CEDAT are bringing the Startup World Cup regional finale to RVITM Bengaluru. Applications close July 10, the top 10 startups are announced July 20, and the regional finale feeds into a Silicon Valley grand finale with a $1M investment prize.
AI Skills Sprint - A 4-Workshop Series for Modern Professionals
A hands-on AI workshop series for working professionals covering AI-ready use-case framing, prompt and workflow design for non-coders, AI-augmented career portfolios, and product/UX decision-making.
HirePro Momentum'26
An invite-only Bengaluru symposium for industry and academia leaders on AI-ready talent, skills-first assessment, high-tech hiring beyond IT, purpose-driven careers, and campus hiring signals through a report and hackathon finale.
eChai Startup Demo Day in Bengaluru
A monthly founder demo session where early-stage product startups show what they are building, get live questions and feedback, and are listed on the event page so attendees can follow up after the room.