Here’s to our community of founders finding their way to the top of US and global AI business!
Founder Spotlight - Christine Ryss:
Christine is building Fanfluencer, where GenAI meets authentic fan feedback from Reddit to craft high-trust marketing video in the AI era. An ex-marketing executive and builder, she also leads Founders Running Club SF - impressive resilience!
And yes, there will be opportunities in Fanfluencer’s Seed round.
Reply to this email if you want an intro ASAP!
VC & accelerators you should know about:
Entrepreneurs First — The Bridge (Bay Area): 8-week US residency helping European founders relocate with visa support and a path to $250k pre-seed.
Deadline: Nov 1, 2025 (Spring ’26 cohort) • Apply
AiSprouts VC — Applied AI Founder Bootcamp: Referral-only, hands-on bootcamp for applied-AI founders (tactics, network, access to capital).
Deadline: Rolling / referral-based • Link
Creator Fund — €41M for European scientific founders: New fund backing university deep-tech (AI, biotech, robotics, materials). Link
Read until the end to find an offer for YOU as a founder or venture scout!
Frontier AI: Thinking Machines (Mira Murati)
The news: Thinking Machines have finally released their first product!
Tinker, is a developer-first API that abstracts distributed fine-tuning and RL on open-weight models (Llama, Qwen).
Now in business language: Tinker is a managed post-training layer for AI models that abstracts the hard stuff into a simple API. It cuts MLOps time and cost, letting teams use first-party data without heavy infra.
Why it matters: Pulls “frontier-style” post-training out of elite labs and into normal research and startup workflows. Time-to-experiment and infra pain both drop.
In July, Murati’s Thinking Machines Labs attracted $2bn in investment valued at $12bn, in a round led by a16z (Andreessen Horowitz), before even bringing a product to market. Other investors included chips giants Nvidia and AMD, as well as Accel, ServiceNow, Cisco and Jane Street.
Signals to watch: (1) Which open weights get first-class support; (2) early case studies with wall-clock + cost metrics. WIRED+1
Future of work: same jobs, more interviews - and better vibe-coding
Walmart CEO Doug McMillon says AI will touch “literally every job” but expects overall headcount roughly flat over 3 years. What? I guess we’d need to look into a different kind of role mix. TechCrunch
Alex raised $17M Series A to automate first-round interviews. Faster funnels, more candidate throughput - but be prepared to replace dozens of CVs with dozens of talks with AI, I guess. TechCrunch
Anything: “vibe-coding” shop claims $2M ARR in 2 weeks and a $100M valuation. The catch? Vibe-coding tool for ready to ship products, not prototypes you wouldn’t know how to scale. Let’s see if it’s the Shopify of vibecoding - or maybe Zapier. TechCrunch
OpenAI → Sora app (with Sora-2): Open AI launched invite-only, TikTok-style feed; tool can generate startling deepfakes (early feeds full of Sam Altman cameos), raising fresh consent + moderation headaches. TechCrunch+1
Link to watch Sora’s launch video and details
Meta have also launched Vibes - short-form AI-video feed inside the Meta AI app. If AI-generated “slop” gets normalized, expect a higher demand for creator authenticity. And proper marking of AI-generated reality, potentially. TechCrunch
Globals & immigration — policy shock hits gig supply
Whizz (e-bikes for couriers) says Chicago riders are fleeing streets amid federal deportation squad news. After scaling to ~300 bikes by July, business dipped ~8% following August raids. Ops risk from enforcement is now material for last-mile platforms and their vendors. And any business employing immigrants, really. TechCrunch
AI monetization models
Result-based billing: Paid raised $21.6M seed to power results-based billing and metering for AI agents. Positioning as the Stripe-for-agents, it’s ready to break seat/credit pricing.
Why it matters: Pricing clarity unlocks adoption; infra players that meter outcomes can become standards. Watch: supported agent frameworks, fraud/chargeback model, early logo momentum. TechCrunch
DeSci spotlight — Periodic Labs
Ex-OpenAI/DeepMind leads raise $300M seed to build “AI scientists” and autonomous labs (first wedge: new materials). If they can close the loop (hypothesis → robot experiments → iterate) and generate proprietary physical-world data, they’ll redefine moats in hard-tech R&D. Is it democratizing science, and making it from university-powered into YC-styled activity? TechCrunch
Zero-friction AI adoption — Composite
Browser-native automation (Chrome extension) that learns and executes multi-step tasks across any site; $5.6M seed from NFDG (Nat Friedman, Daniel Gross), Menlo, and Anthropic’s fund. Why it matters: zero-friction install beats “new AI browsers”; local execution is privacy-friendly—good wedge into agentic knowledge work. Watch: reliability on messy enterprise UIs; policy/SSO hurdles. PR Newswire
Community offer for you:
If you are a venture investor or a startup scout, get an early invite to our network of scouts and VC analysts. 20+ people are already joining to exchange deals in AI, B2B tools, healthcare, and more!
Fill in the form for VCs: https://forms.gle/dgWMrefd5Kuy5haw5
If you are a startup founder, please DM me to be featured in this newsletter. Tell me why you’re amazing - and I’ll share your blurb with 100s of readers of this newsletter (email + LinkedIn), and startup scouts in your niche!
*Feature offer for founders is subject to my (a bit subjective) approval, and limitations of space in this newsletter. So you better be fast!
Fill in the form for founders: https://forms.gle/ZQ2HR5BKgMTA5E4t7
Thank you for reading this issue of Discover with Jane weekly newsletter.
Reply to this email to let me know your feedback, ideas, or anything you’d like to share!
Let’s keep sharing out ventures and adventures!
Jane Kubryak
Social & entertainment — AI video platforms nobody asked for?