Open to 2–5 hrs / bi-weekly Chilean · based in London · GMT github · @enaguero linkedin · /in/enaguero

Hi, I'm Erwin Agüero. A Chilean software engineer based in London who has been teaching, in one form or another, for over a decade.

I'm a Chilean software engineer based in London. Teaching has been part of my work since university — most consistently over the past six years at 4Geeks Academy Spain, alongside engineering roles at Babylon Health, Zappi, and now Hurdle. I'd love to bring that same patience and structure to ByteByteGo's deeper topics.

Email me How I got here Read a sample review I wrote

A small note about this page

I built this site in the open with Claude Code so you could see the kind of work I'd be teaching. The chat further down is a live Claude endpoint grounded in my CV — ask it anything; it will tell you when it doesn't know. The source is public, and every commit that involved Claude carries a Co-Authored-By: Claude line.

Astro 5 Tailwind 4 Claude Sonnet 4.6 Vercel
↗ github.com/enaguero/cv-bytebytego

The teaching arc

How I got here.

Teaching has never been a side project for me — it has been a parallel craft that I've kept returning to, alongside engineering, since university. The bi-weekly cadence at ByteByteGo would let me keep going.

  1. 2008 – 2014

    Teaching Assistant · Pontificia Universidad Católica · Santiago, Chile

    Through my engineering degree I worked as a TA across several courses: Introduction to Programming in Java, Software Engineering (design patterns, agile methodologies), Microeconomics, and Probability & Statistics. This is where I first noticed that helping someone understand a concept I had just learned myself was the part of the week I looked forward to most.

  2. May 2019 – Present · ~6 years

    Full-Stack Bootcamp Instructor (part-time) · 4Geeks Academy Spain · Remote

    I've taught successive cohorts of the full-stack bootcamp — HTML, CSS, JavaScript, React, Python, Flask/FastAPI, SQL, JWT auth, Git. The academy provides the foundation material; on top of that I've built a more structured, comprehensive companion repo for my latest cohort (129) so students leave with a deeper understanding of what they're actually doing — not just a recipe that works.

    Cohort 129 companion repo

  3. 2026 · in progress

    Personal initiative · enaguero.github.io/python-book

    A fundamentals-first Python handbook I just started writing. Two parts (Foundations and Beyond the Basics), short chapters, exercises, and in-browser execution on key topics via Pyodide. It's an experiment — I'll see how it goes — but it's already helping me clarify the pieces I think are worth teaching incrementally.

    Read the work-in-progress

  4. Next chapter

    The role I'm applying for · ByteByteGo · part-time instructor

    ByteByteGo would let me keep teaching while moving into deeper material than a bootcamp — System Design, Agentic AI Coding, AI Automation. The bi-weekly cadence also fits better with my current senior engineering role at Hurdle, and lets me commit to a long-term relationship with the platform.

Why me

A fair question — let me answer it directly.

Honestly, I love teaching. I find helping people understand something new genuinely rewarding — that's been the constant through TA work at university, six years at 4Geeks Academy, and now writing a Python handbook in my spare time.

If you'd like the bullet-point version of what sets me apart from another bootcamp instructor:

Things to look at

Concrete teaching artifacts.

Three pieces of evidence, each linked. Not a portfolio of polished products — actual repos, reviews and notes from teaching.

4Geeks Academy · Cohort 129 companion repo

Lead instructor (Spain, part-time)

A bilingual (Spanish/English) companion repo I built on top of the academy's foundation material, organised day-by-day across 29 sessions. It covers HTML/CSS/JS, React with hooks and routing, Python with Flask/FastAPI, SQL with SQLAlchemy, JWT auth, and Git workflow — with progressive exercises and small projects (Star Wars API client, contact manager, todo apps).

“I prefer the incremental approach in teaching, focusing on the fundamentals as well — so the structure builds one layer at a time, not all at once.”
29 structured daysBilingual (ES/EN)Multiple consumption formats

↗ github.com/enaguero/4geeks_academy_spain_fs_pt_129/blob/main/README.en.md

Formative review of a student's final project

Agentic AI + my pedagogy, worked example

A long-form review for a bootcamp student's capstone (ShadowMap). Claude read the student's repo and produced a first-draft review; I then went line-by-line — correcting misreads, sharpening the framing, and personalising the message — before sending it. It opens by celebrating what's working, separates urgent fixes from polish from excellence, and grades across 11 dimensions with code citations.

“This document is a review of your project designed so you learn, not so you get discouraged.”
11 dimensionsPrioritised action planAI-drafted, human-refined

↗ /sample-review

Python from the fundamentals (work in progress)

Personal teaching initiative · 2026

A Python handbook I'm writing in the open. Two parts: Foundations (computer basics, variables, control flow, functions, OOP, testing) and Beyond the Basics (collections, async, typing). Short chapters, exercises, in-browser execution on key topics. I'll see how it goes — but it's already a useful place to test how I'd structure a longer course.

“Fundamentals first, with examples you can actually run.”
~12 modulesIn-browser exercisesSearch-first reading

↗ enaguero.github.io/python-book/

How I teach

Areas I'd love to teach

Three tracks — with honest framing on each.

I won't claim mastery in areas where I'm still learning. For each track below I've tried to be specific about what I already teach confidently, what I'm actively studying, and what I'd bring as an instructor.

system design

System Design

Teaching what I teach now, learning the parts I haven't taught yet.

At 4Geeks I already teach the building blocks: REST APIs, relational databases, auth, caching, a frontend consuming an API. ByteByteGo's depth — distributed systems, scaling, consistency-vs-availability trade-offs — is material I'm actively studying. I'd love to teach the foundations confidently while learning the deeper layers alongside more advanced learners.

  • Comfortable with: HTTP, REST, relational modeling, JWT, simple caching
  • Currently studying: consensus, replication, partitioning, queue patterns
  • What I'd bring: pacing, runnable examples, honest 'I don't know yet'
agentic ai coding

Agentic AI Coding

I use Claude Code daily and have notes from the failures.

I've been working with Claude Code as a day-to-day collaborator for several months across personal projects and work tasks. This site was scaffolded that way — the commits at github.com/enaguero/cv-bytebytego are honest about what I prompted, what I had to redo, and where the agent went off the rails. I'd teach from that same place: practical workflows, real failure modes, and the parts I'm still figuring out.

  • Day-to-day use across personal and work projects
  • Familiar with Claude Code, MCP tooling, structured prompts and plans
  • Comfortable being explicit about what I still get wrong
↗ github.com/enaguero/cv-bytebytego
ai automation

AI Automation

Start with one repetitive thing. Automate the smallest version of it first.

Most teams I've worked at have a handful of repetitive tasks sitting between engineers and the work they want to do. I enjoy walking learners through how to spot one of those, build the smallest pipeline that solves it, then layer in reliability and observability one step at a time. The CI/CD that publishes this site on every push to main is a fine starting example.

  • Identify the leverage point before reaching for tools
  • Serverless + API glue — Vercel functions, webhooks, scheduled jobs
  • Observability and cost as part of the lesson, not an afterthought

A small live demo

Ask my CV a question.

Rather than just describing the AI tracks, here is a tiny example: the chat on the right is a Vercel serverless function calling Claude with my CV as system context. It's the same pattern I'd use in lesson one of the Agentic AI track — small enough to read in one sitting.

  • About 50 lines of API code, readable in one sitting.
  • The same shape works for docs Q&A, support bots, internal tools.
  • Failure modes — rate limits, hallucination, latency — are part of the lesson, not edge cases.

Try a sample:

POST /api/ask · claude-sonnet-4-6
Hi! I'm a small Claude assistant grounded in Erwin's CV. Ask me about his teaching journey, the tracks he'd like to teach at ByteByteGo, or anything else from his background — and I'll be upfront when something is outside what I know.

Get in touch

If there's a fit, I'd love to talk.

Two to five hours bi-weekly fits well around my current senior engineering role. Happy to send a writing or teaching sample, do a short live walkthrough, or just have a conversation about what your learners need.

Email me · enaguero@gmail.com LinkedIn · /in/enaguero GitHub · @enaguero