Zero Prime Ventures

🎃 October is here and AI is officially in its spooky era 👻 where agents come alive and founders disguise themselves as… other founders? We are hosting a halloween founder gathering. Come dressed as your favorite founder and join us for an evening of great people, good energy, and good vibes. 📅 When: October 30th 📍 Where: Location revealed after registration 🏆 There will be a prize for the best Founder Costume 👀 RSVP below 👇🏻👇🏻👇🏻

SF founders, mark your calendars and get your founder fits ready 🎃. Zero Prime is hosting a Halloween happy hour on Oct 30 and you’re invited! Come ready to talk agentic systems, connect with others building in AI and compete for best costume. Limited spots (founders only). RSVP + approval required. 👉 Register here: https://luma.com/bstceu4w

Companies spent 20 years burying their phone numbers online, training you NOT to call them. AI changed that. Real-time transcription + LLMs just wrapped the phone call in software. OpenAI chairman and Sierra Co-founder Bret Taylor dropped this insight on Decoder: "We've basically put the internet on the phone; we've just made the phone a channel for digital technology for the first time." Voice calls used to be a dead end: $10-20 per call, no logs or triggers, no way to connect to your stack. Most brands' ARPU was too low to justify the expense. Now with transcription and LLMs, voice has new relevance. Here's what that unlocks: → Calls trigger downstream systems like any other API → Voice transcripts become queryable, searchable data → Phone interactions connect to inventory, CRM, payment rails → You can actually measure outcomes, not just handle time Cool concept and worth the listen: https://lnkd.in/gTcJcbdb

Congratulations to Modal on their $87M Series B! Four years ago well before the GenAI wave, Erik and Akshat recognized that cloud infra would need a ground-up overhaul for AI-native companies. They spent those years doing the hard work: building a custom file system, container runtime and scheduler, so teams could focus on writing code instead of managing GPUs or troubleshooting infra. Thousands of teams now deploy in hours instead of weeks and scale cleanly without adding infra overhead. Congrats again to Erik, Akshat, and the entire Modal team.

We're thrilled about Structured AI's announcement on joining Y Combinator. We invested because they're solving real problems for skilled engineers by giving back time lost to endless emails and repetitive tasks. A great example of how AI will augment productivity across industries. Congratulations Raymond Issy Brandon!

Congratulations to Modal on their $87M Series B! Four years ago well before the GenAI wave, Erik and Akshat recognized that cloud infra would need a ground-up overhaul for AI-native companies. They spent those years doing the hard work: building a custom file system, container runtime and scheduler, so teams could focus on writing code instead of managing GPUs or troubleshooting infra. Thousands of teams now deploy in hours instead of weeks and scale cleanly without adding infra overhead. Congrats again to Erik, Akshat, and the entire Modal team.

🎉 Huge congrats to the Wasmer team on launching Python support for Wasmer Edge! As one of their investors, we've watched them tackle one of the hardest problems in developer infrastructure: achieving true sandboxing without sacrificing performance - and they’ve nailed it! It's been great seeing the news float to the top of Hacker News, even with seasoned devs like Simon Willison testing it within hours and sharing feedback. You know you’ve built something special when the community reacts like this! Why developers are going wild: 🏎️ Actually fast - Near-native Python performance in WebAssembly 🔒 Truly sandboxed - Full isolation without sacrificing compatibility 🚀 Zero code changes - Run FastAPI, Django, Streamlit as-is (no Lambda wrappers needed) ⚡ Lightning-quick cold starts - Up to ~6x faster speeds Amazing work Syrus Akbary Nieto and team!

When a 75GB CSV brings everything else to its knees… engineers reach for DuckDB. A Redditor recently shared their struggle: - 75GB CSV, ~400 columns, 2B rows - Dirty data, special characters, duplication - On-prem SQL Server target - Tried SSIS and Python - best case: 8 days to process a single file. (eight!) The “best” reply: “You can query the file with DuckDB if you only need a subset of the columns and it will handle out-of-memory processing. It can query CSVs directly with FROM read_csv('[path]')... and then dump the result to Parquet.” And it worked. The OP processed all their CSVs into Parquet, cut down to the fields they actually needed, and loaded them in a fraction of the time. It’s awesome to see engineers start to default to using DuckDB. When a product becomes the obvious answer in threads like this, it stops being “just a tool” and starts becoming “infrastructure”. DuckDB is doing for data engineers what Postgres once did for app developers: becoming the trusted, go-to backbone for hairy data projects. And this is still the best type of marketing and product market fit: engineers recommending your product to other engineers. (MotherDuck, built around DuckDB, also happens to be a Zero Prime Ventures portfolio company. Seeing the community rally around tech like this is genuinely exciting!)

When a 75GB CSV brings everything else to its knees… engineers reach for DuckDB. A Redditor recently shared their struggle: - 75GB CSV, ~400 columns, 2B rows - Dirty data, special characters, duplication - On-prem SQL Server target - Tried SSIS and Python - best case: 8 days to process a single file. (eight!) The “best” reply: “You can query the file with DuckDB if you only need a subset of the columns and it will handle out-of-memory processing. It can query CSVs directly with FROM read_csv('[path]')... and then dump the result to Parquet.” And it worked. The OP processed all their CSVs into Parquet, cut down to the fields they actually needed, and loaded them in a fraction of the time. It’s awesome to see engineers start to default to using DuckDB. When a product becomes the obvious answer in threads like this, it stops being “just a tool” and starts becoming “infrastructure”. DuckDB is doing for data engineers what Postgres once did for app developers: becoming the trusted, go-to backbone for hairy data projects. And this is still the best type of marketing and product market fit: engineers recommending your product to other engineers. (MotherDuck, built around DuckDB, also happens to be a Zero Prime Ventures portfolio company. Seeing the community rally around tech like this is genuinely exciting!)