Google’s SIMA 2 agent uses Gemini to reason and act in virtual worlds

✨TodayOnAI’s Daily Drop

• Google’s SIMA 2 agent uses Gemini to reason and act in virtual worlds

• Milestone raises $10M to make sure AI rhymes with ROI

• LinkedIn adds AI-powered search to help users find people

• 💬 Let’s Fix This Prompt

• 🧰 Today’s AI Toolbox Pick

Google

Google’s SIMA 2 agent uses Gemini to reason and act in virtual worlds

🚀 TodayOnAI Insight: Google DeepMind has previewed SIMA 2, its next-gen embodied AI agent powered by Gemini 2.5, marking a major step toward general-purpose AI. Unlike its predecessor, SIMA 2 not only follows instructions—it understands environments, reasons through tasks, and learns from experience.

🔍 Key Takeaways:

• SIMA 2 integrates Gemini’s language and reasoning to double SIMA 1’s performance, handling more complex tasks in unseen environments.

• The agent can self-improve using AI-generated tasks and feedback, learning behaviors with minimal human data—similar to how humans learn from trial and error.

• Embodied interaction is central: SIMA 2 perceives, reasons, and acts within 3D games and photorealistic worlds, such as those created by DeepMind’s Genie model.

• Demonstrations included emoji-based instructions (🪓🌲 → chop a tree) and contextual reasoning (e.g., identifying a “ripe tomato” red house).

• No release timeline yet; the preview aims to spark collaborations and future real-world applications, including potential in robotics.

💡 Why This Stands Out: SIMA 2 showcases how language models and embodied intelligence can converge into agents that don’t just follow commands—they understand context, adapt, and evolve. As AI shifts from chatbots to agents that can navigate and reason in dynamic environments, SIMA 2 offers a glimpse at what AGI might require in practice. How soon will such agents cross from simulation to real-world robotics?

Milestone AI

Milestone raises $10M to make sure AI rhymes with ROI

🚀 TodayOnAI Insight: Israeli startup Milestone has raised $10M to help enterprises measure the ROI of generative AI in software development, offering granular insights into code quality, productivity, and team efficiency—all powered by access to customers’ codebases and tool stacks.

🔍 Key Takeaways:

• Purpose-built analytics: Milestone’s platform correlates generative AI tool usage with engineering metrics like feature delivery speed and bug attribution.

• "GenAI data lake" model: Combines data from codebases, project management tools, team structure, and codegen platforms to offer actionable insights.

• Backed by key players: Raised a $10M seed round led by Heavybit and Hanaco Ventures, with participation from Atlassian Ventures and GitHub co-founder Tom Preston-Werner.

• Enterprise-first strategy: Focuses solely on large organizations, with current customers including Kayak, Monday.com, and Sapiens.

• Academic roots, global team: CTO and Trinity College professor Stephen Barrett co-founded the startup remotely with his former student, CEO Liad Elidan.

💡 Why This Stands Out: Milestone addresses a critical blind spot in today’s AI-driven engineering stacks: the inability to measure how—and how well—generative AI is actually being used. By building infrastructure that ties usage to outcomes, it helps teams move beyond hype to hard data. As engineering teams shift from writing code to orchestrating AI agents, tools like Milestone may become essential to managing that transition.

LinkedIn

LinkedIn adds AI-powered search to help users find people

🚀 TodayOnAI Insight: LinkedIn is rolling out AI-powered natural language people search to U.S. premium users, aiming to simplify how members discover relevant connections. The move marks a significant upgrade to one of the platform’s most-used features—shifting away from filters and keyword matching to conversational queries.

🔍 Key Takeaways:

• LinkedIn now supports natural language search for people, expanding beyond its earlier AI-powered job search tool.

• Users can type prompts like “Find me investors in the healthcare sector with FDA experience” or “Who in my network can help me understand wireless networks.”

• The feature is currently available to U.S.-based premium users, with broader availability planned.

• Traditional filters and exact-title searches are no longer required, addressing a long-standing user pain point.

• While early results show promise, LinkedIn notes that query interpretation still has inconsistencies it's actively refining.

💡 Why This Stands Out: AI is fundamentally reshaping how we access professional networks, and LinkedIn’s update places it in direct competition with emerging AI-native platforms. By allowing natural language queries, LinkedIn reduces friction in people discovery—an area ripe for disruption. As AI assistants increasingly act on our behalf, will platforms like LinkedIn remain open, or follow Reddit in tightening data access?

✨ See how a simple prompt upgrade can unlock better AI output.

🔹 The Original Prompt

At first glance, this prompt might seem okay. But it's too broad — and that limits the quality of AI-generated results. Let’s improve it using prompt engineering best practices.

✅ The Improved Prompt

💡 Why It's Better

• Specific audience: Targets decision-makers in mid-sized companies.

• Contextual focus: Emphasizes emerging tech and practical solutions.

• Actionable output: Requests summaries and titles to spark execution.

• Tone and style: Guides the type of content (insightful, engaging, relevant).

🛠️ Learn how to adapt this prompt for SaaS, AI tools, dev teams & more →
Read the full PromptPilot breakdown

💡 Bonus Tool: Want to generate and master prompts instantly?
👉 Try PromptPilot by TodayOnAI (Free to use)

📰 More from the AI World

• Court rules that OpenAI violated German copyright law; orders it to pay damages

• Anthropic announces $50 billion data center plan

• Microsoft-backed Veir is bringing superconductors to data centers

• The global race for the AI app layer is still on

🧰 Today’s AI Toolbox Pick

• 🍋LemonSqueezy (Finance Tool): Handles the tax compliance burden so you can focus on more revenue with less headache.

• 💻ZipWP (Web Design Tool): Creates stunning websites in seconds.

• ⚙️DupDub (Content Tool): An all-in-one content creation platform that allows you to craft your content effortlessly and streamline your workflow.