Engineering at @kiva
Medellin, Colombia
Joined December 2014
- Tweets1.9K
- Following1K
- Followers679
- Likes13.9K
Esteban Vargas retweeted
Great engineer 10 years ago: Knowing 1 programming language and framework.
Great engineer today: A tech-agnostic tech lead with good problem-solving abilities and people skills.
Esteban Vargas retweeted
Something I have been thinking about: in the past, the best engineers I knew spent a lot of time automating their work in various ways. Better vim/emacs automations, writing lint rules to catch repeat code issues, building up a suite of e2e tests so they don't need to smoke test the app manually. These kinds of things were the highest leverage activities an engineer could do, because it multiplied their own output, which in turn meant they could build more things.
I think many of these automations have become even more important now. This is true for a number of reasons.
First, infra and DevX automation speeds you up. And if you are running an army of agents, each of those agents will be sped up also. More automation == more output per unit of time.
Second, moving things to code improves efficiency. Your agent could fix an issue every time it sees that issue happen, but that uses tokens and might miss cases. If Claude instead writes a lint rule, CI step, or routine, that class of issue can be fully automated forever. This is really what people are talking about when they talk about loops -- it's about automating entire types of busywork rather than solving them one off. This isn't a new idea at all. Engineers have been doing this for a long time!
Third and most importantly, automation makes it possible for others to contribute to the codebase more easily. Increasingly what I am seeing is engineers are contributing to codebases on day one because Claude can navigate the codebase for them, and that non-engineers are able to contribute to a codebase as effectively as engineers can. What gets in the way of both of these is domain knowledge that lives in peoples' heads rather than in automation -- the stuff you used to have to learn when ramping up. What has changed thanks to agents is the domain knowledge that can be encoded as infrastructure is no longer limited to what is expressible in lint rules and types and tests; it can now capture nearly all domain knowledge, encoded as code comments and skills and CLAUDE.md rules and memories. If I put up a PR for an iOS codebase I don't know and a code reviewer rejects it because it doesn't use the right framework, or if a designer builds a new feature and it gets rejected because it doesn't follow the right architectural patterns, these are failures of automation.
Every team should be writing the CLAUDE.md's, REVIEW.md's, skills, and docs that enable agents to productively work in their codebase with zero additional context from the prompter. This sounds crazy, and at the same time is a natural extension of the stuff engineers have always done: automate, and encode domain knowledge as infrastructure. As the model gets smarter and as the harness matures, this task becomes easier. In the meantime, it is on every team to look for ways to convert their domain knowledge to infra so that Claude can write code better, so that code review catches issues automatically, and so the next person working on your codebase can contribute more easily.
Esteban Vargas retweeted
Imagine a world where we simplify every role in a company down to just two:
1. Builders
2. Sellers
Esteban Vargas retweeted
Before vibe coding became a thing, programming was already evolving in that direction. It already increasingly consisted of installing and configuring stuff other people wrote, without reading the source.
Esteban Vargas retweeted
Software hiring has become absurd.
At work, you’re expected to use AI to offload manual coding and move faster.
Then, to get your next contract, you’re asked to code from memory with no assistance.
Pass the interview - and you’re expected to use AI again.
It has never been this broken.
Esteban Vargas retweeted
As an AI Engineer. Please learn
>Harness engineering, not just prompt engineering
>Context engineering, not just long prompts
>Prompt caching vs. semantic caching tradeoffs
>KV cache management, eviction, reuse, and memory pressure at scale
>Prefill vs. decode latency and why they optimize differently
>Continuous batching, paged attention, and throughput optimization
>Speculative decoding vs. quantization vs. distillation tradeoffs
>INT8, INT4, FP8, AWQ, GPTQ, and when quantization hurts quality
>Structured output failures, schema validation, repair loops, and fallback chains
>Function calling reliability, tool contracts, argument validation, and idempotency
>Agent guardrails, loop budgets, tool budgets, and termination conditions
>Model routing, graceful fallback logic, and degraded-mode UX
>RAG architecture: chunking, embeddings, hybrid search, reranking, and freshness
>Retrieval evals: recall, precision, grounding, attribution, and citation quality
>Evals: golden sets, regression tests, adversarial tests, LLM-as-judge, and human evals
>LLM observability as a first-class discipline: traces, spans, tokens, latency, errors, and drift
>Cost attribution per feature, workflow, tenant, and user journey not just per model
>Safety engineering: prompt injection defense, data leakage prevention, and permission boundaries
>Multi-tenant isolation, cache safety, and cross-user context contamination prevention
>Fine-tuning vs. in-context learning vs. RAG vs. distillation and when each is the wrong tool
>Latency, quality, cost, and reliability tradeoffs across the full inference stack
>Production failure modes: hallucinated tool calls, malformed JSON, stale retrieval, runaway agents, and silent eval regressions
Esteban Vargas retweeted
Seré el Presidente de Colombia con la ayuda de Dios y el Pueblo el próximo 21 de junio; en mi plan de gobierno he planteado establecer la asociación de internet rural más grande del mundo con @SpaceX. @elonmusk ha sido líder mundial combatiendo la pobreza conectando a la gente en línea en cualquier rincón del mundo; el siglo XXI no se puede empezar a vivir sin conexión para todos los colombianos.
Le plantearé a @elonmusk una alianza de conectividad y que conozca a Colombia y lo que puede ofrecer para el desarrollo de su revolución tecnológica.
Firme por la Patria. 🫡
(A.D.L.E) 🇨🇴🐅
Esteban Vargas retweeted
Something strange is happening in tech.
CTOs of billion dollar companies are quitting to take IC roles at Anthropic.
Workday CTO -> MTS (Mar 2026)
You[.]com CTO -> MTS (Mar 2026)
Instagram CTO -> MTS (Jan 2026)
Box CTO -> MTS (Dec 2025)
Super[.]com CTO -> MTS (July 2025)
Adept AI CTO -> MTS (Jan 2025)
The mission is that real.
Esteban Vargas retweeted
Very excited to share (a preview of) @stripe projects:
A new way for you—and your agents—to programmatically provision and manage the internet’s best developer services, right from the CLI.
projects.dev to sign up for early access.
Demo ⤵️
Esteban Vargas retweeted
Issue tracking is dead.
We are building what comes next.
linear.app/next
Esteban Vargas retweeted
#2 is important, so I built it. Keeping you accountable with other people pushes you forward
Write your weekly standup, vote on other people's, climb up on the leaderboard.
Looking forward to contine building this at the Cafe Cursor in Buenos Aires! personal-standup.vercel.app/
Esteban Vargas retweeted
Stripe CEO Patrick Collison: "Software should be like pizza… cooked right then and there at the moment of use."
"You don’t want mass-produced industrial scale software. You want bespoke custom software made for you, that moment."
"Up until now, the economics of software have been conceived as fixed cost and then infinitely monetized."
"Once there are inference costs and custom creation involved, it really shifts. It’s kind of the non-Walrasian software regime."
@patrickc with @collision on @tbpn
Esteban Vargas retweeted
Bought a new Mac mini to properly tinker with claws over the weekend. The apple store person told me they are selling like hotcakes and everyone is confused :)
I'm definitely a bit sus'd to run OpenClaw specifically - giving my private data/keys to 400K lines of vibe coded monster that is being actively attacked at scale is not very appealing at all. Already seeing reports of exposed instances, RCE vulnerabilities, supply chain poisoning, malicious or compromised skills in the registry, it feels like a complete wild west and a security nightmare. But I do love the concept and I think that just like LLM agents were a new layer on top of LLMs, Claws are now a new layer on top of LLM agents, taking the orchestration, scheduling, context, tool calls and a kind of persistence to a next level.
Looking around, and given that the high level idea is clear, there are a lot of smaller Claws starting to pop out. For example, on a quick skim NanoClaw looks really interesting in that the core engine is ~4000 lines of code (fits into both my head and that of AI agents, so it feels manageable, auditable, flexible, etc.) and runs everything in containers by default. I also love their approach to configurability - it's not done via config files it's done via skills! For example, /add-telegram instructs your AI agent how to modify the actual code to integrate Telegram. I haven't come across this yet and it slightly blew my mind earlier today as a new, AI-enabled approach to preventing config mess and if-then-else monsters. Basically - the implied new meta is to write the most maximally forkable repo and then have skills that fork it into any desired more exotic configuration. Very cool.
Anyway there are many others - e.g. nanobot, zeroclaw, ironclaw, picoclaw (lol @ prefixes). There are also cloud-hosted alternatives but tbh I don't love these because it feels much harder to tinker with. In particular, local setup allows easy connection to home automation gadgets on the local network. And I don't know, there is something aesthetically pleasing about there being a physical device 'possessed' by a little ghost of a personal digital house elf.
Not 100% sure what my setup ends up looking like just yet but Claws are an awesome, exciting new layer of the AI stack.
Esteban Vargas retweeted
FAANG software engineer tells how they vibe code at FAANG
---
"You still always start with a technical design document. This is where a bulk of the work happens. The design doc starts off as a proposal doc. If you can get enough stakeholders to agree that your proposal has merit, you move on to developing out the system design itself. This includes the full architecture, integrations with other teams, etc.
Design review before launching into the development effort. This is where you have your teams design doc absolutely shredded by Senior Engineers. This is good. I think of it as front loading the pain.
If you pass review, you can now launch into the development effort. The first few weeks are spent doing more documentation on each subsystem that will be built by the individual dev teams.
Backlog development and sprint planning. This is where the devs work with the PMs and TPMs to hammer out discrete tasks that individual devs will work on and the order.
Software development. Finally, we can now get hands on keyboard and start crushing task tickets. This is where AI has been a force multiplier. We use Test Driven Development, so I have the AI coding agent write the tests first for the feature I’m going to build. Only then do I start using the agent to build out the feature.
Code submission review. We have a two dev approval process before code can get merged into man. AI is also showing great promise in assisting with the review.
Test in staging. If staging is good to go, we push to prod."
---
reddit. com/r/vibecoding/comments/1myakhd/how_we_vibe_code_at_a_faang/
Esteban Vargas retweeted
Tech debt is like other debt:
- Longer you hold it, more it costs
- Costs accumulate over time
- Pay it off ASAP
- Sometimes worth taking on debt, to do something important faster
- Some take on debt without thinking: can I pay it off
- Taking on more debt to pay off debt is bad