Founding GTM Engineer at Saturday Robot

I run Saturday Robot, a fractional head-of-growth practice for Series A companies. I help founders turn early wins into repeatable growth systems across positioning, messaging, outbound, and content. I'm hiring a junior, scrappy GTM engineer to work alongside me: someone who builds the systems and writes the words that run through them.

Why work with me

AI and market forces in general are changing GTM in three ways.

Your job is to create systems, not campaigns. Repetitive work that used to be the anchor of many early roles is being trivialized by AI agents steered by good prompts and taste. It's harder than ever to build that foundation of knowledge to be great at it.
You need to create offers as good as products. It's SO noisy. Your offer needs to create enough value for your prospects to cut through. An offer absolutely cannot be "jump on a sales call to hear about how you can pay me money." It needs to be just as engaging and well packaged as the best content in their feed.
Tactics reserved for the largest enterprise accounts are now moving downmarket. AI makes it economical and necessary to map the buying committee, track intent signals, multi-thread across several people, and coordinate outreach across channels for every meaningful account.

What I'm offering is the opportunity to learn how to be world class across all three by working with companies just setting the foundation after their Series A/B.

My work is to help founders transition from one-off sales or campaigns to creating their very first growth system. You'll help me build that for them.

The engineering only counts if the writing is good. A sequence that reaches exactly the right 200 people still gets ignored when the first line is weak, so you'll spend as much time on the offer, the subject line, and the reply as on the enrichment that picked the list. Same for how you talk to people: you'll be in threads with founders and VPs, and in the client Slack every week.

A bit of background

LinkedIn ↗

I've spent my career across growth, marketing, and product for B2B SaaS companies, as both a leader and a founder. I led growth at Kite, an AI coding assistant from before ChatGPT that reached 7M developers before it was acquired. At Affirm, I led growth for merchant top of funnel and the mobile app.

I founded Firezone through YC W22, and for a brief moment it was Reddit and HN's favorite corporate VPN. Most recently I ran GTM at GrowthX, where we grew to $12M in 9 months building AI content engines for companies like Lovable, Surge AI, Ramp, and Vercel.

A few of the companies I've led growth at or worked directly with:

A few things that show how I think and work:

What you'll do

Map the TAM for each client. Find every company that could plausibly buy. Then work out which of them are worth a message this quarter, and why.
Build the audience from that map. Enrich each account with the signals that say it is in the market now. Keep the data clean as the list grows.
Write what goes out. That is the offer, the sequence, the page it points to, and the reply that keeps a conversation alive.
Stand up the sending infrastructure. You will set up domains, configure DNS records, warm mailboxes, set daily limits, and watch deliverability.
Build the tooling we do not have. AI writes most of the code now, so a scoring script, a reply classifier, or a small internal app is a day of work rather than a quarter.
Own the follow-through. Check the live campaigns, read the replies, and move the conversations that deserve a real answer. Keep notes clean enough that the client can read them.
Improve the playbook. Try new tools, write down what worked, and make the next client's setup faster than the last one.

Requirements

You are comfortable with data. You can dedupe a list built from three sources and normalize the fields. You will notice when a merge quietly drops 400 rows. SQL helps and is not required.
You can read API documentation and get a working request out of it. That means auth headers, JSON, pagination, and rate limits. You do not need a CS degree.
You build with AI. You can get a working internal tool out of an AI coding tool and debug it when it breaks. You can also point a model at a few thousand records to sort them, then check a sample before trusting the result.
You automate work instead of repeating it. A script, an n8n flow, or a scheduled job all count. You also build a way to know when it breaks.
You know how email sending works, or you can learn it in a month. Sending domains, SPF, DKIM, and DMARC should stop being acronyms. Mailbox warmup, daily limits, and list validation decide whether any of it arrives.
You write well. About a fifth of this job is writing, and it has to be your work rather than a model's first draft. You will also be in threads with founders and VPs, and in a client Slack every week.
You finish work that arrives without instructions. Point at something you built, shipped, or sold where nobody handed you the steps.

What a typical day looks like

Map a client's TAM, then cut it down to the accounts worth contacting this week. Enrich those accounts with the signals that explain why now.
Stand up the infrastructure a campaign needs: domains, mailboxes, sending limits, tracking, and the checks that catch a broken run before it goes out.
Write and ship the next campaign, then read the replies closely enough to understand what the response is telling us.
Design an offer ladder: what someone gets for a click, what they get for a reply, and what they get for a call.
Automate a manual step, or build the small app that does it. An afternoon of work with an AI coding tool often replaces a recurring hour.
Look at pipeline and audience growth per client, and say plainly which system is producing and which one is not.

Practical details

20-30 hours per week, with room to go full-time.
Remote.
Series A B2B companies, mostly AI products.

How to apply

Send me the four things below. In the last box, tell me about yourself, something you built or pushed through, and why this role interests you.