engineer and explorer - starting something new
Los Gatos, CA
Joined September 2009
- Tweets877
- Following358
- Followers745
- Likes1.7K
Jonah Back retweeted
Lots of engineers think AI codegen is only good enough to do little bug fixes here and there
If you tell them you can ship 15k LOC of ai gen code to prod they think you have lost your mind.
It’s so so early. And also those engineers are living in 2025.
Jonah Back retweeted
I’m not exaggerating, I hear from so many big software cos which don’t use Claude Code/Codex.
CTOs are asleep at the wheel. Engineers are typing code by hand. Fixing a bug a day. Like it’s 2024.
If youre at these cos, demand change or leave. Now. You’re in for a rude awakening.
Jonah Back retweeted
It’s time to build in San Francisco again
WSJ today on the San Francisco real estate market
Pretty good summary what’s going on, and it’s not overly complicated:
“An uptick in demand coupled with the city’s notorious lack of housing supply means that fierce bidding wars are breaking out again for single-family homes and condominiums in desirable neighborhoods”
Jonah Back retweeted
The cruelest thing you can do to an engineer is require them to use AI to code and at the same time restrict them to using crappy models and tools.
I understand why it happens but would be so frustrating.
Amazon had four Sev-1 outages (their highest severity level) in a single week. Internal memos say AI-assisted code changes were a contributing factor.
The timeline here is wild. In October 2025, Amazon laid off 14,000 corporate employees. In January 2026, another 16,000. That’s about 30,000 people in five months, roughly 10% of the corporate workforce. CEO Andy Jassy said the cuts were about culture, not AI.
During those same months, Amazon set a target: 80% of developers using AI coding tools at least once a week. They tracked adoption closely and blocked rival tools like OpenAI’s Codex. Even so, 30% of developers still hadn’t touched Amazon’s in-house tool Kiro by January.
In December 2025, Kiro caused a 13-hour AWS outage. The AI tool had production-level permissions and decided the best fix for a bug was to delete and recreate an entire live environment. A second incident involved Amazon Q Developer, another AI tool. Amazon blamed both on “user error, not AI.” But quietly added mandatory peer review for all production access afterward.
Then March 5: Amazon’s retail site went down for about six hours. Over 22,000 users reported checkout failures, missing prices, and app crashes. Amazon called it a “software code deployment” error.
Five days later, SVP Dave Treadwell made the normally optional weekly engineering meeting mandatory. His memo acknowledged “GenAI tools supplementing or accelerating production change instructions, leading to unsafe practices.” These problems trace back to Q3 2025. Amazon’s own assessment: their GenAI safeguards “are not yet fully established.”
The new rule: junior and mid-level engineers now need senior sign-off on any AI-assisted production changes. Treadwell also announced “controlled friction” for the most critical parts of the retail experience.
For context, Google’s 2025 DORA report found 90% of developers use AI for coding but only 24% trust it “a lot.” An Uplevel study of 800 developers found Copilot users introduced 41% more bugs with no improvement in output. Amazon is finding out what those numbers look like at the scale of a $500 Billion revenue company, with 30,000 fewer people on staff to catch the mistakes.
Jonah Back retweeted
Most data teams are still building dashboards nobody looks at.
We killed 1,400 of ours.
Replaced them with a semantic layer you can talk to. Built custom AI tooling that replaced SaaS we were paying six figures a year for. Every person on this @Opendoor data team has unlimited tokens and AI agents running alongside them daily.
Real estate is the largest asset class on earth and the transaction still feels like 1997. That's the data problem we wake up to. Pricing millions of homes. Optimizing tens of millions in ad spend. Building models that move real dollars, not slide decks.
Our CEO's (@nejatian) mandate is "default to AI." Not a slogan. How we actually operate. Opendoor might be the most AI-native public company you haven't been paying attention to.
Small team. Absurd talent density. Embedded across the entire C-suite. You touch pricing, marketing, product, sales, operations — all of it!
We're hiring across Agentic Analytics Engineering and Data Science. Seattle or Toronto.
If this is the kind of data team you've been looking for, stop scrolling and apply:
Agentic Analytics Engineering: grnh.se/mqzo4ab46us
Data Science: grnh.se/awisi8t36us
Jonah Back retweeted
Last shirt said "Faster." Sneak peek at what this one says. Limited drop at openarmy.com now powered by @shop.
Don't miss Q4 Financial Open House tomorrow 2/19 at 2pm PT on X, Robinhood, and YouTube. investor.opendoor.com
Jonah Back retweeted
For anyone in a software engineering leadership/management position, this is the time to see if you can figure out how to shelve that responsibility for a while and return to being a full time IC.
Software engineering has changed more in the last few months than it has in the previous decade. The skills, experience, and perspective that you built to get you to the leadership/management position you're in now are becoming irrelevant extremely rapidly. If you are not able to adapt right now, you are going to start becoming bad at your job very soon, if you have not already.
You're trusted to lead because you built the skills and judgement over time direct the boat confidently. But the boat that you were driving has suddenly been replaced by a completely different boat which you have never worked on before, so it's time to shift down and re-build your expertise before taking the helm again.
🫡
Jonah Back retweeted
Or @Opendoor.
Use all the tokens you could possibly want.
At YC, we let our engineers let it rip. There is no limit to the number of tokens you get.
If you're at a place that is stingy with tokens, and you have that need for a speed, come to YC and let it rip.
ycombinator.com/software
Jonah Back retweeted
Client wanted multi-region active-active. I talked them out of it.
Their situation:
- 2,000 daily users, all in Western Europe
- 99.9% uptime requirement
- 18-month runway
- 3 engineers total
What they actually needed:
- Single region with proper HA
- Multi-AZ database with automated failover
- CDN for static assets
- Solid backup and restore tested monthly
Cost difference:
- Multi-region estimate: $23,000/month
- Single region HA: $4,200/month
They hit 99.95% uptime last year. Zero regrets.
Not every problem needs a distributed systems solution. Sometimes the answer is just good fundamentals done well.
Jonah Back retweeted
Hire out of pain. Don't hire because you think you'll need someone soon or maybe sometime later. Wait until you or your team are actually hurting: working weekends, missing family dinners, dropping balls. That pain is the signal that the role is real. I learned this the hard way after watching founders (including myself) hire ahead of need and end up with people in roles that weren't fully formed yet. When you hire out of pain, you know exactly what the job is because you've been doing it yourself. You can evaluate performance because you know what good looks like. And the new hire knows you'll step back in if they fail, because you were just doing it last week.
Jonah Back retweeted
yes things are changing fast, but also I see companies (even faang) way behind the frontier for no reason.
you are guaranteed to lose if you fall behind.
the no unforced-errors ai leader playbook:
For your team:
- use coding agents. give all engineers their pick of harnesses, models, background agents: Claude code, Cursor, Devin, with closed/open models. Hearing Meta engineers are forced to use Llama 4. Opus 4.5 is the baseline now.
- give your agents tools to ALL dev tooling: Linear, GitHub, Datadog, Sentry, any Internal tooling. If agents are being held back because of lack of context that’s your fault.
- invest in your codebase specific agent docs. stop saying “doesn’t do X well”. If that’s an issue, try better prompting, agents.md, linting, and code rules. Tell it how you want things. Every manual edit you make is an opportunity for agent.md improvement
- invest in robust background agent infra - get a full development stack working on VM/sandboxes. yes it’s hard to set up but it will be worth it, your engineers can run multiple in parallel. Code review will be the bottleneck soon.
- figure out security issues. stop being risk averse and do what is needed to unblock access to tools.
in your product:
- always use the latest generation models in your features (move things off of last gen models asap, unless robust evals indicate otherwise). Requires changes every 1-2 weeks - eg: GitHub copilot mobile still offers code review with gpt 4.1 and Sonnet 3.5 @jaredpalmer. You are leaving money on the table by being on Sonnet 4, or gpt 4o
- Use embedding semantic search instead of fuzzy search. Any general embedding model will do better than Levenshtein / fuzzy heuristics.
- leave no form unfilled. use structured outputs and whatever context you have on the user to do a best-effort pre-fill
- allow unstructured inputs on all product surfaces - must accept freeform text and documents. Forms are dead.
- custom finetuning is dead. Stop wasting time on it. Frontier is moving too fast to invest 8 weeks into finetuning. Costs are dropping too quickly for price to matter. Better prompting will take you very far and this will only become more true as instruction following improves
- build evals to make quick model-upgrade decisions. they don’t need to be perfect but at least need to allow you to compare models relative to each other. most decisions become clear on a Pareto cost vs benchmark perf plot
- encourage all engineers to build with ai: build primitives to call models from all code bases / models: structured output, semantic similarity endpoints, sandbox code execution. etc
What else am I missing?
I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last ~year and a failure to claim the boost feels decidedly like skill issue. There's a new programmable layer of abstraction to master (in addition to the usual layers below) involving agents, subagents, their prompts, contexts, memory, modes, permissions, tools, plugins, skills, hooks, MCP, LSP, slash commands, workflows, IDE integrations, and a need to build an all-encompassing mental model for strengths and pitfalls of fundamentally stochastic, fallible, unintelligible and changing entities suddenly intermingled with what used to be good old fashioned engineering. Clearly some powerful alien tool was handed around except it comes with no manual and everyone has to figure out how to hold it and operate it, while the resulting magnitude 9 earthquake is rocking the profession. Roll up your sleeves to not fall behind.
Jonah Back retweeted
software development in 2026 is going to require some to loosen up a little
code doesn't have to be as perfectly crafted the way we did it pre-ai
call it slop if you want, but if you're still demanding perfection on every pr while your competitors are shipping "slop" that works... you're fighting from a disadvantaged position
shipping velocity matters more than perfection
Jonah Back retweeted
Great minds! We actually just went live with a big test that does exactly this. 🎯
It’s about transparency: showing the math for Cash vs. Cash Plus in one view so sellers can choose the path that fits their goals. All of this is enabled by our powerful & recently more flexible OVM, our AI-driven nationwide pricing engine. Sellers can now easily pick the option and cost structure that works for them. Our "Cash Now, More Later" path is a total no-brainer for convenience. We want sellers to see the math and move with 100% confidence. 🏡 $OPEN
See if you can spot the new design (they show up after assessment) in the wild. We're always eager for feedback.
Jonah Back retweeted
First pic took 10 *years* of work without AI.
Second pic took 10 *weeks* of work with AI.
ZIP codes where @Opendoor can buy homes.
Jonah Back retweeted
Proud of the engineering rigor behind this expansion. Beyond the coverage, the team achieved a 90% reduction in model sensitivity issues and a 50%+ improvement in pipeline runtime. It’s rare to see a launch that increases scale while simultaneously driving down costs and latency. Incredible work by the OVM team.
Jonah Back retweeted
Excited to have @lucmatheson join Open from Coinbase as our President. Let’s go!
$OPEN Opendoor Names President and CFO to Accelerate Next Chapter investor.opendoor.com/news-r…
