Studio
About
Daniel Hull is the founder of 80x, an agentic-engineering studio for venture-capital funds.
What the studio is
80x is an independent studio: one person plus agents. Dan writes the specifications and reviews everything that ships; AI agents do much of the building, testing, and monitoring. An agentic-engineering studio, in plain terms, is one that builds systems where AI agents do real work inside your tools, under controls that keep them safe to run unattended.
What it ships
- Client systems inside Attio and Affinity: qualification agents, migrations and cleanups, and the scheduled automations behind fund dashboards. The results are documented in the work.
- Open documentation of the studio's methods at 80x.ai/docs.
- Open-source tools, listed at 80x.ai/tools and published on GitHub.
- One commercial product: Backchannel, a native client that syncs WhatsApp and LinkedIn conversations to Attio or Affinity.
How the studio works
Four habits appear in every engagement:
- Spec-first builds. Every system starts as a written specification you can read and correct before any code runs.
- Dry-run by default. Anything that touches your data first runs in a mode that reports what it would change without changing it.
- Proof-gated autonomy. An agent earns the right to act unattended only after its output has been checked against results known to be correct.
- Numbered decision logs. Every judgment call made during a build is recorded under a number, so you can trace why the system behaves the way it does.
Why the docs are open
The methods are published so buyers can check the work. Before you hire the studio you can read exactly how it plans a migration, gates an agent's autonomy, and logs its decisions in the docs, then compare that against the case studies in the work. Nothing has to be taken on faith.
Where to go next
- The docs — how the systems are built.
- The work — case studies with real numbers and anonymized clients.
- The tools — open-source tools and Backchannel.
- GitHub — the public repositories.
- daniel@80x.ai — email Dan directly.