All Stories — Page 116
Every story published on RuntimeWire, newest first.
- Rangeview wins Kratos turbine-engine work for flight-quality superalloy castings
Cameron Schiller's El Segundo foundry will develop flight hardware as Kratos scales engines for drones and missiles.
- Startup Spotlight: Atomarine launches gas-first floating data-center plan with nuclear later
The YC S26 company wants to ship 75-100 MW compute barges by 2028, then replace gas power vessels with marine reactors.
- Wiz says Azure Cosmos DB flaw could have exposed every database
Wiz says Microsoft fully remediated CosmosEscape, a Gremlin API vulnerability chain that reached a platform-wide signing secret.
- Lamina Labs raises $4.8 million to turn AI answers into explainer videos
The two-person YC startup's Simi product turns prompts and documents into narrated whiteboard videos for people, software and AI agents.
- Arena launches AutoEval to rank new AI models within an hour
The reward-model system uses millions of user preferences and scored above 0.98 correlation with Arena's later live rankings.
- OpenAI cuts GPT-5.6 Luna API prices 80% and adds faster Sol tier
Luna falls to $0.20 input and $1.20 output per million tokens; Sol Fast mode promises up to 2.5x speed at twice the standard rate.
- Moonshot releases Kimi K3 report, detailing how a 2.8T open model runs
The July 27th report details hybrid attention, 16-of-896 expert routing and a serving stack that still demands data-center scale.
- Simile raises $200M at a $2B valuation to simulate human behavior
Greenoaks led the round five months after the Stanford spinout emerged with a $100M Series A.
- Firecrawl says its rebuilt MCP cuts context use by 50%
The web-data service added OAuth onboarding for people and keyless access for agents, reducing setup friction across MCP clients.
- Scale AI appoints Francis deSouza CEO for enterprise AI push
The former Google Cloud COO takes over on August 10th as Scale AI expands its applications business across enterprises and governments.
- Lucebox partners with AMD on a $6,499 local AI inference box
Founder Alessandro Puppo says the Radeon R9700-Strix Halo system ran DeepSeek V4 Flash 3.63x faster than one DGX Spark.
- Xsight Labs raises more than $300 million for open AI networking chips
Fidelity led the round at a $2.8 billion valuation as Xsight Labs scales programmable Ethernet switches and 800G DPUs.
- Applied Electrodynamics unveils WaveSight to map pipes and wiring through walls
The YC S26 hardware maker says WaveSight images hidden building systems from across a room, with construction as its first market.
- Microsoft releases EvoLib for LLM agents that learn without weight updates
The framework turns model trajectories into reusable skills and insights, with research code released under an MIT license.
- Google adds system-wide voice editing to Gemini's Mac app
The Fn-key shortcut can dictate polished text or use screen and file context to rewrite selections, summarize documents and insert images.
- Dili discloses $15M Series A for capital-project compliance AI
Khosla Ventures led the financing, which closed roughly six months before Dili's July 30th announcement.
- K2 Space raises $500 million at $6.8 billion valuation
Brothers Karan and Neel Kunjur more than doubled K2 Space's valuation after the Gravitas mission reached orbit and SES outlined an initial 28-bus plan.
- Nscale agrees to buy Ray startup Anyscale for reported $1.65 billion
Anyscale would add a software layer to the vertically integrated AI infrastructure stack Josh Payne outlined before Nscale's rapid capital buildout.
- MIT founders launch Atomarine to build gas-first nuclear data centers at sea
Dimitris Koutentakis and Emile Germonpre target a gas-powered pilot by 2028, with marine reactors planned as a later swap.
- MiniMax teases H3 as a unified multimodal model for Hailuo AI
MiniMax is framing H3 as a single system for text, images, video and sound, aimed at carrying creative context across media.
- Intropy raises $11 million to automate spare-parts decisions in ERPs
Felix Capital led the seed round for Intropy, founded by former Tractable researchers Franziska Kirschner and Yih Kai Teh, as the AI spare-parts software maker prepares a New York office.
- Tesla produces its 10 millionth vehicle, six years after crossing 1 million
The milestone car rolled out of Fremont as Tesla prepares the California factory for a larger role in robotics.
- UCSB and LinkedIn researchers train AI agents to predict their own tool calls
The approach raised exact-match accuracy on five benchmarks while reusing one model and its KV cache, though real-world latency gains remain unproven.
- X launches encrypted Chat API for bots and AI agents
The open-source SDK supports six languages, while X is offering pay-per-use accounts a temporary allowance of 500 messages per day.