Eric Lamanna
A leader at the DEV.co family of brands — including Automatic.co — Eric focuses on turning agentic AI from a demo into dependable, production-grade automation that teams actually trust.
- Agentic-AI delivery leadership
- Workflow-to-fleet roadmaps
- Governed, human-in-the-loop autonomy
- Secure, production deployments
Pragmatic about autonomy
Eric spends his time where strategy meets the messy reality of shipping software.
Eric works across the DEV.co family of brands — dev.co, seo.co, llm.co, and Automatic.co — where the through-line is the same: take a real, high-leverage workflow and make it run reliably without constant human babysitting. He is far more interested in the workflow that quietly saves a team ten hours a week than in the agent that looks impressive in a demo and breaks the moment it touches a production system.
Most of his conversations start with a simple question — where does work actually get stuck? — and end with a mapped architecture, clear guardrails, and a measured path to production. He treats autonomy as something you earn step by step, with approval gates and decision lineage in place before you hand an agent the keys to anything that matters.
He is plain-spoken about trade-offs, allergic to hype, and happiest when the answer is a smaller, sharper system that a team can understand, operate, and trust.
Where Eric spends his time
A few of the problems he and the Automatic.co team work on every week.
Agent orchestration
Coordinating multiple specialized agents around a single workflow, so the right step runs at the right time with the right context.
Earned autonomy
Designing systems that take on more responsibility as trust and ROI compound — never all at once, never unsupervised by default.
Human-in-the-loop
Building approval gates and exception handling so people stay in control of the decisions that carry real risk.
Systems integration
Connecting agents to the ERPs, CRMs, databases, and APIs where work actually lives — safely and observably.
Governance & lineage
Risk thresholds, model governance, and end-to-end decision lineage that hold up to scrutiny and audit.
Private deployments
Designing for the customer's perimeter — on-prem, VPC-isolated, or air-gapped — from the first conversation.
Assess before you automate
Eric's default move is to slow down at the start. Before anyone writes production code, he wants the workflow mapped, the failure modes understood, and the human checkpoints agreed on. That up-front clarity is what lets the build move fast later without nasty surprises.
The goal is never automation for its own sake. It is a system the team can reason about — one where every action an agent takes is logged, reversible where it needs to be, and traceable end to end.
- Map the workflow and its failure modes first
- Agree on human checkpoints up front
- Log every action with full decision lineage
Have a workflow worth automating?
Bring a real problem and Eric and the Automatic.co team will help you map the path from workflow to a governed agent fleet.