Phosphor Capital

CONFIDENTIAL. Do Not Share. Delve 7.0 coming soon.

gumloop raised a $50m series b led by benchmark here's a video we had fun making about the journey back to work.

Wrapping up the week with a heartfelt message from one of our long-time users. When you're deep in the weeds building agents and watching dashboards all day, it's easy to forget what it actually feels like on the other side. This is why we do it. Happy Friday.

@BrainbaseHQ this is brilliant! well done team!

We're excited to release Brainbase Conversational, the first ever code-runtime general agent designed for perfectly reliable and traceable voice agents.

@the1024th well done team! proud to be backing you!

Gauge is your marketing agent for organic, paid, and AI search.

@MaxBrodeurUrbas love the new branding and this video! well done on announcing series b - we're so proud to be backing you!

@_adishj impressive launch!

3/3] why this matters this launch represents a significant moment for us as a product. ui is melting away and development and engineering work is more focused on the orchestration layer. instead of forcing you to learn a new ui, we want to meet you and your agents where you already are. our goal is a full-stack, omnipresent video infrastructure that runs wherever you are. this api launch is the first step in that direction.

today, we are making the @mosaic_so video editing api available to all agents & humans. see how we setup larry — our slack openclaw agent — with a mosaic api key and had it clip, edit, and post archived steve jobs footage. all without ever leaving slack. you can connect your openclaw agents, claude code, or vibe-coded saas to our agentic video editing api. no waitlist — now live at mosaic [dot] so. comment "API" to get 100 free credits dropped into your account to get started with the api for free. more details on why & how we’re making this change below (thread):

@haakamaujla @generalcatalyst Congrats @haakamaujla and team! Excited to back this one alongside @generalcatalyst .

@sumanyu @HammingAI @ycombinator @livekit @pipecat_ai @Vapi_AI @retellai Well done on the launch — 5M+ calls monitored speaks for itself! 📷 Proud to be backing you!

@HammingAI  (@ycombinator S24) is the first complete QA platform for voice agents from pre-deployment testing to production monitoring. Voice agents break in production in ways you can't predict. Hamming connects to your agent (@livekit, @pipecat_ai, @Vapi_AI, @retellai, @elevenlabs, @usebland, or any custom stack), auto-generates test cases from your prompt, and runs 1,000+ parallel calls over phone, SIP, or WebRTC. Failures are flagged with exact root causes. Regressions are caught before deployment. 10K+ agents + 5M+ calls monitored. 95% agreement with human evaluators. First test report in under 10 minutes.

Worked at Tesla. Now I'm building a wire harness factory The difference between Tesla and the other OEMs? The engineers actually think about what they're designing. Everyone else just copies what they did last time with a few tweaks. Had a customer using a super expensive multi-conductor cable. Asked why. "That's what the engineer before me used." Switched them to single conductors. Literally 90% cheaper, works the same. We made less money but they became more competitive, can continue to build cool shit and hopefully grow the pie of atoms that go to market.

Today, we announce Tamarind Bio’s $13.6M Series A, led by @_DimensionCap, with participation from @ycombinator. Tamarind has now become the trusted platform for molecular AI inference, serving tens of thousands of scientists, including 8 of the top 20 pharma, dozens of biotechs and research organizations. Since last year, we’ve grown revenue 7x and are grateful to work with world-class biopharma R&D organizations like @Bayer, @Boehringer, and Adimab. — When we started two years ago, a few brave biotechs took a chance with us. Hoping this new company, providing easy access to computational drug discovery tools, would enable their scientists to apply foundation models to real therapeutic applications. Since then, our library of AI models has grown to hundreds, not just open-source tools, but also our users’ internal protocols, proprietary models trained on users’ own data, pipelines of multiple models together and more! Now, we are doubling down on our commitment to building the core AI and data infrastructure to power the next generation of medicines. We will continue to support open models, strengthening them with proprietary data, and prioritize access to all scientists. Join us as we build the infrastructure for all AI-powered drug discovery.

At Delve, we’re 3% the team size of our competitors doing 30% of their monthly revenue. I can confidently say that companies created in the last 3 years look completely unrecognizable from the inside compared to companies created 5+ years ago.

@ycombinator @proliferateai Big problem. Great team. Excited to be backing Proliferate.

Local coding agents can't do end-to-end work at your company. @proliferateai is the open-source platform for company-specific agents that can. Provision secure access to your tools, set up automations, and let agents work in the background. Congrats on the launch, @thepablohansen! ycombinator.com/launches/PVw-p…

🚨: Scientists mapped 1 mm³ of a human brain ─ less than a grain of rice ─ and a microscopic cosmos appeared.

@yuvansundrani @autosanaai @ycombinator @DeVC_Global @468Capital Big problem. Great team. Congrats on the raise 🙌

excited to share that @autosanaai raised $3.2m in seed funding to reimagine how software is tested. thanks to @ycombinator, @DeVC_Global, @468Capital , and amazing angels including paul graham (founder of yc), eric levine, kulveer taggar, and several of our customers. we felt this pain firsthand: manual qa takes days, bugs still ship, and e2e frameworks are brittle + expensive. autosana closes the testing loop fast. get set up: autosana.ai press release in comments.

@aquariusacquah Rebuilding your entire product 2 weeks in takes serious conviction. This team ships. Proud to back them

2 weeks ago, we rebuilt our entire product. "Browser automation" fell short of our mission to eliminate all repetitive knowledge work. The new Kaizen is the ultimate digital employee: always on, extremely capable, continually learning. Sign up for access in the tweet below.

We just raised $6.3M seed to help dev teams build and deploy MCP Servers for AI Agents and MCP Apps for ChatGPT / Claude Our open source SDKs crossed 4M downloads when we recorded this video (now already 6M+). Our mission is to build tools that enable humans to build software for agents, as agents will be the primary consumers and manipulators of digital information. We make building with MCP effortless. Create MCP servers or MCP Apps with mcp-use SDK. Preview and iterate with mcp-use Inspector. Connect your GitHub repo to our Manufact Cloud. Instantly make your MCP Server available to billions of AI agents and 800M+ users of ChatGPT / Claude.

@manufact congrats on the raise! excited to back this high output team.

@Marcel7an @usetusk We love what you're building at @usetusk.

Introducing Tusk 2.0. Tests pass. AI code review looks good. But your product still breaks. The future of code review is a deterministic tool that tests changes against real user behavior. @usetusk turns your production traffic into a self-maintaining API test suite. Cover thousands of real-world edge cases without writing a test again.

We're excited to partner with @perplexity_ai on their latest release. We were impressed with how performant their new Deep Research product was on early benchmarks and are thrilled that this work is being open sourced.