Field notes

The ledger.

15 Free GitHub Repos Every Claude Code Setup Needs

Claude Code is powerful out of the box, but the right open-source repos make it much better at planning, remembering context, testing its own work and connecting to the rest of your engineering stack.

Ghita EL Haitmy · Aug 28, 2026

15 Free GitHub Repos Every Claude Code Setup Needs

ResourcesAug 18, 2026

12 Signs Your Company Is Vibe Spending (And What Each One Costs You)

Most companies can name every employee on payroll. Very few can name every AI tool, model or agent their teams are already paying for. This guide breaks down vibe spending into 12 signs you can check this week, with the formulas to calculate what each one is actually costing you.

Ghita El Haitmy

ResourcesAug 18, 2026

What Happens When One AI Agent Can Access Every App?

AI agents used to sit inside one tool. Now, one agent can read your inbox, update your CRM, check an invoice, push code and book a meeting — all from a single prompt. That is powerful. But it also creates a problem most scale-ups have not planned for: Who owns the agent, what can it access and how much is it really costing you?

Ghita El Haitmy

ResourcesAug 14, 2026

I paid $5 for a $0.50 coding task

The problem was not the answer. It was who was doing the reading. I had a premium model opening files, searching references, reading logs, and carrying all of that raw context inside the main conversation. Now I split the work. A cheaper subagent reads and searches, while the stronger model gets the findings and handles the part where I actually need stronger reasoning. This works in Claude Code and Cursor.

Ghita El Haitmy

ResourcesAug 11, 2026

The Leaky AI P&L

AI rarely blows a budget in one dramatic decision. It leaks through thousands of individually reasonable ones, and most companies still budget it like ordinary SaaS instead of the variable, usage-driven cost it has become.

Ghita EL Haitmy

ResourcesAug 10, 2026

Why Is AI So Hard to Price? AI Pricing Models in 2026

AI is hard to price because nobody agrees on what creates the value: access, usage, work performed, or results. Seats made sense for software built around people, but one AI agent can do thousands of tasks without a seat, and the alternatives to seat pricing (usage, credits, outcomes) bring their own problems. A 2026 study of 631 finance, product and engineering leaders found only 8% were fully confident they knew what their AI features cost to deliver, showing how many companies are pricing before they understand their own costs.

Ghita El Haitmy

ArticleJul 28, 2026

Tech Startup Funding Rounds: Last 30 Days, Updated Weekly

The AI funding rounds keep coming, month after month, and the checks aren't getting smaller. This is where we're tracking July through December 2026: who raised, who wrote the check, and what each bet actually says about where AI is headed next. Less hype, more receipts.

Ghita El Haitmy

ArticleJul 20, 2026

5 ways to Cut Your LLM Bill Without Cutting Quality👇 (save this)

If you're a founder, COO, or Head of IT at a scaleup right now, your AI spend probably isn't one line item anymore. It's scattered across agents, chatbots, unapproved vendor subscriptions, and unexplained API bill jumps, with no system to make that spend legible or controllable. The fixes are well-documented by Stanford, Microsoft, and Amazon, but nobody owns applying and enforcing them company-wide. This piece isn't code to paste in, it's what to ask your engineering team and why your AI spend owner needs visibility across every vendor and agent before these fixes actually save money. Save this. Send it to whoever owns your AI infra budget.

Ghita El Haitmy

ResourcesJul 17, 2026

Sapiom vs ELI (2026): Do You Need an Agent Execution Engine or Agent Spend Governance?

Sapiom is an execution engine that lets AI agents buy and run services; ELI governs what your AI agents spend. Here is how the two compare in 2026 — and when to use each.

Ghita El Haitmy

ArticleJul 16, 2026

Understanding AI ROI

Companies are burning money on scattered AI subscriptions and agent usage with zero tracking, and the moment an investor asks about ROI, there's no real answer. This piece explains why that gap exists and how to close it before it becomes a board problem.

Ghita El Haitmy