Operating Plan · Learn AI project

The Build-to-Master Track

A 26-week execution plan for going from beginner coder to someone who ships real AI products. Not a syllabus to admire — a contract with yourself.

Start: beginner coder Target: build & ship AI products ~20+ hrs/week Math: decent foundation ~26 weeks intensive Synced to Google Calendar
Overall progress 0 / 0 tasks

Here's the honest version. You are a beginner coder who wants to build and ship AI products, with real time to put in — 20-plus hours a week is close to a part-time job. That is enough to get genuinely good in six months, but only if you spend it building, not collecting courses. The single most common failure mode at your stage isn't lack of talent, it's tutorial hell: watching lecture after lecture, feeling productive, and having nothing to show for three months except notes. This plan is built to make that failure mode structurally hard to fall into — every phase below ends with something running, deployed, or public, not just "I understand it now."

You have a decent math foundation already, so this plan doesn't stop to re-teach you calculus from zero. It picks up the linear algebra and probability you need exactly when a phase needs it, and no earlier — math for builders is a tool you sharpen on the way to the job, not a gate you have to clear first. The same discipline applies everywhere: depth is earned by what the current project demands, not by what looks impressive on a roadmap.

Six months of consistent building will put you meaningfully ahead of almost everyone who "learns AI" by consuming content. It will not make you a research scientist, and it doesn't need to — your goal is products, not papers. Hold yourself to the checklist, not the vibe of feeling like you're learning.

Non-negotiables

  1. 01Ship something runnable every two weeks. A script, a demo, a deployed app — no exceptions, no "still polishing."
  2. 02No new course or tutorial starts until the current one has produced a working project you can point someone to.
  3. 03Keep a build log — a few lines a day on what you built and what broke. This is how you catch yourself drifting into passive learning.
  4. 04Every phase ends with something public: a GitHub repo, a deployed link, a short write-up. Private notes don't count as done.
  5. 05If a topic isn't required to ship what you're currently building, skip it and note it for later. Depth comes from repetition across projects, not front-loading theory.
0
Foundations
Weeks 1–3 · Aug 24 – Sep 13, 2026 · ~65 hrs

Python fluency and real developer tooling: git & GitHub, the terminal, virtual environments, an editor you're fast in. Enough programming fundamentals — functions, data structures, file I/O, error handling — that code stops feeling like magic.

DeliverableThree small working scripts pushed to GitHub (e.g. a file organizer, a simple web scraper, a CLI tool that calls a public API), plus a personal notes repo you'll keep adding to.
CalendarPhase block Aug 24–Sep 13 · review checkpoint Sun Sep 13, 8pm
0/5
1
APIs, Data & Your First AI App
Weeks 4–5 · Sep 14 – Sep 27, 2026 · ~40 hrs

REST APIs, JSON, the `requests` library, pandas basics for wrangling data, and just enough front-end (Streamlit or Gradio, not raw HTML/CSS/JS) to put a UI on top of your code. Then: your first call to an LLM API and your first deployed AI app.

DeliverableProject 1 — a small deployed app that calls an LLM API to do something useful (e.g. summarize a document you paste in, or answer questions about a CSV you upload).
CalendarPhase block Sep 14–27 · review checkpoint Sun Sep 27, 8pm · Ship Day Sun Sep 6 & Sep 20
0/5
2
Math for ML & Classical Machine Learning
Weeks 6–9 · Sep 28 – Oct 25, 2026 · ~80 hrs

Linear algebra and probability, but only the parts ML actually uses: vectors, matrices, dot products, gradients, distributions, Bayes' theorem — applied immediately, not proved in the abstract. Then classical ML with scikit-learn: regression, classification, clustering, and the discipline of proper train/validation/test evaluation.

DeliverableProject 2 — an end-to-end classical ML project on a real dataset (data cleaning → model → evaluation → a simple deployed demo of the result).
CalendarPhase block Sep 28–Oct 25 · review checkpoint Sun Oct 25, 8pm · Ship Days Oct 4 & Oct 18
0/5
3
Deep Learning & How LLMs Actually Work
Weeks 10–14 · Oct 26 – Nov 29, 2026 · ~100 hrs

Neural networks from the ground up in PyTorch, then transformers, attention, and tokenization — enough that "how does an LLM work" stops being a black box. The goal isn't to train GPT from scratch, it's to never again treat a model like magic when you're debugging your own product.

DeliverableA small neural net trained end-to-end in PyTorch, plus a working miniature GPT built by following Karpathy's build-from-scratch series — reproduced yourself, not just watched — with a short written explainer of how attention works, in your own words.
CalendarPhase block Oct 26–Nov 29 · review checkpoint Sun Nov 29, 8pm · Ship Days Nov 1, 15 & 29
0/5
4
LLM Engineering — Building Real AI Products
Weeks 15–19 · Nov 30, 2026 – Jan 3, 2027 · ~100 hrs

This is the phase that actually matters for your goal. Rigorous prompt engineering, retrieval-augmented generation (RAG) and vector databases, tool/function calling, agents that take multi-step actions, evaluating LLM outputs properly instead of vibes-checking, and a first fine-tune.

DeliverableProject 3 — a deployed RAG app that lets someone chat over a real set of documents. Project 4 — an agent that uses tools to complete a multi-step task on its own.
CalendarPhase block Nov 30–Jan 3 · review checkpoint Sun Jan 3, 8pm · Ship Days Dec 13 & Dec 27
0/5
5
Ship Product #1 — Real Users
Weeks 20–24 · Jan 4 – Feb 7, 2027 · ~100 hrs

Everything before this was preparation. Pick one product idea, constrain the decision to a week, and build an MVP using what you now know. Deploy it, get it in front of strangers, and iterate on real feedback — not more building in isolation.

DeliverableA live AI product with its own URL and landing page, at least 10 real users who are not friends or family, and a written feedback log with what you changed because of it.
Resources"The Mom Test" for talking to users without fooling yourself, Product Hunt and Indie Hackers for launch and distribution — no new technical courses this phase, this is execution.
CalendarPhase block Jan 4–Feb 7 · launch retro Sun Feb 7, 8pm · Ship Days Jan 10 & Jan 24
0/5
Deepen, Scale, Repeat
Week 25+ · From Feb 8, 2027 · ongoing

Open-ended by design. What you build here depends on what Product #1 actually revealed you're missing — that's more honest than guessing now. Likely candidates: production-grade evals and monitoring, cost and latency optimization, fine-tuning open models yourself, multi-agent systems, voice or vision. The loop from here is simple: build, ship, learn, repeat — and revisit this plan monthly to be honest about where you actually are.

0/3

If you fall behind — and at some point you will — the fix is never to skip a project to catch up on the calendar. Cut scope inside the current project until it fits the time you actually have, ship the smaller version, and move on. A shrunk deliverable that shipped beats a full one that's still "almost done" in your head. The one thing you're not allowed to cut is shipping itself.

Twenty-six weeks assumes real, defended hours — not hours that evaporate into browsing AI Twitter and calling it research. Protect the time the same way you'd protect a job you actually wanted.