Daniel Hull GitHub ↗

Agents for venture capital

I build agents and tools for venture funds. Software that reads a fund's own data and does a slice of the work a partner or analyst would otherwise do by hand. They stay small, run on the fund's own keys, and the general parts are open source.

For VCs · runs on your keys · daniel@80x.ai

Book a call Send a message

flagship · building now

Valentine

The agent I'm building now. It saves you the embarrassing moment: walking into a call with a founder your fund already met and passed on, with no idea it happened. Before the call, Valentine reads your CRM, notes, email and calendar and tells you who spoke to them, when, and what they decided. It only reads, never writes. Runs on your own token. Works with Attio and Affinity today.

tryvalentine.com ↗ GitHub ↗ npx valentine-agent acme.com

agent · private beta

dealscout

Keeps the MEDIC qualification fields current in a sales CRM, without the data entry. It reads each deal's call notes, pulls out the qualification signals, and suggests them in Attio — every value cited back to the source note, and never overwriting what a rep already entered. There's a read-only Slack bot too: @mention a deal and get its MEDIC picture for call prep. Runs on your own keys.

dealscout-site.vercel.app ↗ private beta

tool · open source

Attio CLI

Drives an Attio CRM from the command line: bulk reads, migrations, dedupe, attribute audits. I use it on every engagement, and it's what my agents call when they need to touch a CRM.

github.com/80x-djh/attio-cli ↗

chrome extension

Slack Channel Member Exporter

Exports any Slack channel's member list to CSV. Slack has no built-in way to do this. I needed it for sourcing work, so I built it.

Chrome Web Store ↗

youtube · building in public

@danieljhull

I post about building this stuff: agents, the tools around them, and what's actually working with private capital data. Mostly so the firms I work with can see how it's made.

youtube.com/@danieljhull ↗

How I work

Most engagements start with the data. Funds sit on years of deals, contacts and notes that are duplicated, half-filled, or spread across tools. I get that into shape first, then build agents on top that are narrow enough to trust against a live CRM. The projects above are the pieces general enough to share.

Working with a fund on this, or just want to compare notes? Book a call or send a message.

Proud to have worked with