Zaigo builds custom AI systems for founder-led, middle-market companies — the kind of businesses with messy operations, real revenue, and legacy software in place. They are thinking about how to leverage AI but don't know where to start. We turn manual workflows and inefficiencies into production AI systems, fast.
We are hiring an AI Implementation Lead to own client delivery end-to-end inside a pod structure. You will run 5–10 client accounts at once, with 2–3 senior AI engineers reporting into your pod, and you will be the person the client trusts to get things shipped.
Think of this as a principal-level role at a consulting firm: a senior partner sets the strategy and owns the relationship at the top, you own the execution and the day-to-day client trust, and your engineers do the heavy build. You are the person who keeps everything moving, makes the calls, and ships.
What you'll actually do
Own the client. You'll run discovery calls, scope the work, set expectations, give status updates, manage scope creep, and be the steady hand the client calls when something is on fire. Many of our clients are non-technical executives running real businesses — you need to be able to translate \"we have this messy thing happening in our operations\" into a concrete AI system, in plain English, without jargon.
Architect the solution. On a discovery call, you should be able to sketch out a working solution in real time: \"Okay, this is a RAG problem with a chat layer on top, we'll use these 5 specialized agents to handle X, Y, and Z, output formatted to your Excel template, and we'll integrate with your existing CRM via API calls.\" You don't need to write the production code, but you need to know enough to confidently scope it, pick the right framework, point to the right open-source repo, and know roughly what's hard vs. easy.
Run the pod. Translate the scoped solution into engineering tasks, hand it off cleanly, unblock the engineers, and review the work. You can't read every line of AI-generated code — nobody can anymore. You're evaluating whether the output is directionally right, whether the UX makes sense, whether the architecture will hold up, and whether it actually solves the client's problem. Strong product intuition matters more than line-by-line review.
Prototype when needed. You're a player-coach. You should be comfortable opening Cursor or Claude Code to validate an approach, build a quick demo for a client call, or prototype a tricky workflow before handing it to engineers. You're not shipping production code as your main job, but you can build.
Juggle. You'll have ~10 active projects across 5–10 clients at any time. Deadlines slip, priorities shift, clients change their minds. You stay calm, communicate proactively, and keep everything moving.
Who you are
You have at least 5–8 years of experience in some combination of technical product management, solutions engineering, technical consulting, forward-deployed engineering, or running your own thing. You may have been a founder. You may have been a PM at a fast company who got tired of waiting for engineers and started vibe coding your own stuff.
You're technically literate in a hands-on way: you can vibe code, you know the major agentic frameworks (Claude Agent SDK, Vercel AI SDK etc.), you understand RAG architectures, you know when to use a vector DB vs. a graph DB vs. just dumping things into context, you've shipped things using AI tools.
You communicate exceptionally well in English, both written and verbal. You're comfortable on a Zoom call with a 62-year-old CEO who has never heard of Loom and explaining to him, patiently, what we're building and why it matters for his business.
You have founder-grade ownership. You don't say \"that wasn't in the ticket.\" You don't wait for someone to tell you what to do next. You see the gap, you close it.
This role is for you if
You can take a messy business problem and produce a clear, shippable architecture for it within a single call.
You like running a lot of things at once and you don't drop balls.
You're excited about AI-assisted building and you use these tools every day already.
You want to work alongside experienced founders and senior engineers, with real ownership and real client exposure, in a fast-growing company.
This role is not for you if
You want to write production code all day.
You're a pure project manager who needs a PRD and a Jira board to function.
You're uncomfortable on the phone with non-technical executives.
You need a manager to tell you what to prioritize this week.
You avoid ambiguity.
Location
Fully remote. We're focused on candidates based in Brazil with strong daily overlap with U.S. Eastern Time.
How to apply
To apply, please email (not DM) the following directly to and copy :
* Your resume
* A short Loom video (5–10 minutes max, must be a Loom link) walking us through one technical or operational project you personally owned
In the Loom, cover:
* The business problem you were solving
* Your specific role
* The technical workflow or architecture at a high level
* The biggest challenge you solved
* How you organized and managed delivery
We strongly prefer candidates who can explain complex technical systems clearly, concisely, and in plain English.
* Applications without a Loom video will not be considered.