Hacker News new | past | comments | ask | show | jobs | submit | shinkim0914's comments login

Eraser | https://eraser.io/ | Full-time | San Mateo, CA | ONSITE | Software Engineer

At Eraser, we're building the copilot for technical design. We're the leader in AI-powered diagramming and have built a platform that includes proprietary diagram-as-code implementation, a free-form infinite canvas, and a markdown-compatible document editor used by hundreds of thousands of users every month.

Interested? We'd love to hear from you. Please send your resume or LinkedIn and a brief explanation of why you're excited about this role to hn@eraser.io.


Hey I am not sure if I am just someone who happened to see your role a little too early when looking to apply for the role, however when clicking the career section, it says there are no job posting as of currently. Perhaps it may just simply be due to some type of glitch and not bearing very much significance, however I would still like to reach out to you about the role and establish a connection as I see myself being highly congruent with the team. If you would like, take a look at my portfolio website. Hope you find it useful in being in gaining at least some type of idea on my expertise and how I would fit in the role.

https://portfolio-sand-xi-26.vercel.app/

You can also take a look at my github repo through one of the shortcuts in my website, as I have also spent a great deal of my time pursuing many open software projects and offering contributions.


I'm the founder of a startup that makes a diagramming software for engineering team called Eraser.

If you want a codified set of rules that you can mechanically apply, I would look at UML and C4. However, I'd argue that the frameworks are overkill for the vast majority of engineering diagram use cases.

For the general scenario where you're just trying to communicate a technical concept clearly without spinning your wheels a lot, here are a few tactical things I'd recommend:

1. Pick the right diagram type for the job. Sequence diagrams are great for API documentation, ERDs are great for database schemas, flow charts are great for user flows, data flows and logical flows.

2. Use icons where possible. A database icon, a S3 bucket icon, a cylinder shape are all a lot more expressive and easier to grok than a rectangle with text inside that says "database".

3. Focus on the nodes, groupings, and connections. That's really all there is to a diagram if you break it down. If you just make a list of nodes, groups, and connections that you want to represent, the diagram will almost just draw itself.

4. If you're looking for a quick way to start jump start a diagram or looking for inspiration, try AI tools to see how far they can get you.


Eraser, https://eraser.io/ looks awesome. I signed up but had to try a few times with the Passwords (mine was a 42-character with all the jazz generated by 1Password).


I never figured out how to make a valid password.


1. Yeah.

2. Strong disagree. IMHO icons in diagramming are obsolete for most purposes (smacking of UML and over-wrought database diagrams) and are likely to lead to confusion for the majority of viewers who - at best - aren't quite sure what you exactly mean in this particular case.

3. Which is what graphviz is for. It literally draws itself. Subgraphs are the groupings.

4. I would instead suggest modifying an existing graphviz .dot file, or man dot.


Re: 2 I would like a large icon and small text that confirms my intuition about what the icon means. The icon thereby eventually becomes part of my lexicon.

I don’t even see the text, I just see database, load balancer, Redis cache…


> Generate DDL directly generate SQL statements to create your database tables.

Looks like it's capable of turning a diagram to SQL too? That's neat!

For those that are interested in an GPT-powered tool, we have a project that can generate diagrams from code of any language or natural language: https://eraser.io/diagramgpt


Hi HN - Excited to share DiagramGPT. Diagrams are helpful to have but painful to draw. What if you could generate diagrams from what you already have? Like code or project descriptions?

• Code → Diagrams

• Natural language → Diagrams

Would love to get feedback and happy to answer any questions.

https://eraser.io/diagramgpt


I do foresee pricing pressure ahead for them as a lot of competitors seem to arising.


Examples of such competitors please? Very much thanks in advance...


There is no place like Silicon Valley where taking a career risk to quit a job and start a company is normalized, understood, and encouraged.


Additionally, California is one of the only states that doesn't recognize non-competes [1], that is key for innovation and competitors coming up including small/medium competitors. This part is always overlooked.

> A few states, such as California, Montana, North Dakota, and Oklahoma, totally ban non-compete agreements for employees, or prohibit all non-compete agreements except in limited circumstances. [1]

There is also a massive augmented wave coming that is heavily underestimated and will change everything. The future is heavily content creation in new phases of technology which are huge. Overall, it is better to have a virtual economy that uses less resources than a physical one.

The new markets are definitely remote and that is how you communicate with most people now even in the same building, so being physically in California isn't as needed. Though the policies of not recognizing non-competes needs to go nationwide. Non-competes are anti-innovation, anti-worker, anti-business, anti-competition and only help the bigs.

[1] https://en.wikipedia.org/wiki/Non-compete_clause#United_Stat...


Tons of companies that aren't in California don't have non-competes or have very limited ones. I've never had a meaningful non-compete in the course of my career. (Had a very narrow one when one company was acquired.)

I'm certainly not a fan of non-competes but note that, even in California, a company can drag you into an expensive court battle over non-solicitation clauses, NDAs, etc. It's also a matter of non-competes not being enforceable in general. A small company may still choose not to hire you if they think there's a possibility they may need to go to court.


I've had 3 in a row in Europe. I doubt they would have been enforceable, but from conversations with colleagues they certainly had a chilling effect.


Yet with the high rents and the lack of universal healthcare it takes a crazy person to take such a risk. I personally will start my startup from abroad when the time is right.


Yes, it’s still risky. But at least you’re still understood because of the pervasive startup culture in SV. Everyone in SV can relate to startups. I think what’s tough outside of SV is that people don’t relate to startups. People might think you’re just weird.


Who cares if your investors are from sv but you’re based elsewhere?


The investors care. Traditionally they wouldn't fund anyone who wasn't in easy reach.

It's also easier to hire employees since they're all in SV.


I am curious. What else do investors "traditionally" care about? I am referring to aspects that aren't always explicit, like physical proximity.

Do investors not also want (competent) exceptions that stand out?


Investors want to meet a slightly smarter but younger version of themselves. This makes them feel comfortable with the risk of their investment, they imagine all kinds of congruencies between you and them, even if they don't exist. They want to convey some nugget of wisdom that you, the entrepreneur, reverence as they key to their success. But most of all, they want you to be a money tree that buds and flowers and bears fruit continually, to the degree the generated revenues are a problem. Anything less, and they go silent for a while before applying pressures with rarely helpful advice from non-technology or old-technology backgrounds.


Considering the amount of random cryptocurrency being funded, heh


Some VCs want to see you have a good team that works together, and some combination of good ideas and work history that makes it seem like your startup will take off even if you have to pivot and give up your current prototype.

Others are having a midlife crisis and want to be your new rich dad, or they want a cult leader who makes them feel smart and throws cool insider parties. In this case it helps to be a white guy or at least Elizabeth Holmes.

One would hope YC is the first since we’re on their website, but having read pg’s essays and noticed his advice for startups is half post hoc fallacies (“use Lisp because I did”) and half is unethical (“don’t hire women or people with accents to get culture fit”) I dunno.


Employees are all in SV? What?


There's a strong distance effect, where investors will invest more easily companies located near them. It could be that this will change after a year of experience with doing everything online, but it's been true historically.


It kinda balances out when you're young and can easily get a seed $1M dollars to build a company with a very dubious (or non-existent) business plan

There's your salary. If the company works out or not later, well, you're a founder and has the experience, it's easier to get a job anywhere.


> Yet with the high rents and the lack of universal healthcare it takes a crazy person to take such a risk.

It takes a rich person to take such a risk, but yay, meritocracy!


he says that the question is what kind of company one will build


... that's because it's not a career risk when you're a multi millionaire who will just be rehired by your friends if your venture fails.


Who are these multi millionaires? Clearly I know the wrong people in tech in the Bay Area.


Probably most people I work with in their 30s or 40s are multimillionaires.


Yeah? On paper, 401k, or elsewhere? Where have they worked and for how long?

Because in my experience, the “everyone who works at a FAANG is a multi-millionaire” meme is 100% bullshit. Especially when I read things like https://news.ycombinator.com/item?id=26336029


I never said everybody is, but I do think most people who have been at FAANG or companies that pay in the same ballpark like trendy late stage pre-ipo companies for 10+ years are or are close (especially if we define multimillionaire as >2m networth including 401k and house equity, and include household wealth for married couples).

My other comment you linked I still stand by. But you don't need to be IC6/L6/E6, refreshers and equity appreciation are very powerful if you are lucky enough to get both.


What company uses the scale in this comment? (IC5 IC6 IC7 etc)


At least Apple does. I'm not sure if other companies use it as well.


It's not the Apple scale though. I mean, it looks like it, but the details are different.


Definite humblebrag here.


Maybe you should check what some of their houses are worth lol


The people I know in tech in the Bay Area rent. It’s their parents, who bought their properties in the 1980s or before, who are the land-owners.

That said, I’d definitely be interested to see actual statistics about own vs. rent for people in tech in the Bay.


If your talented and fail at starting your own company it won’t be very hard to find a job in sv


If you get an entry level TC of 140k and save 50% of it, you'll be a millionaire in 15 years.

That's without any kind of growth, investment, stock options growing in value, raises, etc.


And if you make $65,000 a year and save 100% of it, you’ll be a millionaire in 15 years.

Nobody is arguing that it’s possible.


Whoops, missed a word. I meant not possible.


I don't think that's the primary reason, nor as common as you imply. The old saw about how SV is a place where failure isn't disrespected is closer to the truth.


Why doesn’t that happen elsewhere?


If this sets them on a self-sustaining path without having to rely on running highly conspicuous donation campaigns on Wikipedia, I think that's a wonderful thing.


Not so fast.

Imagine Wikimedia Enterprise becomes the #1 source of revenue for Wikimedia. Shortly after, people will see that Wikimedia is doing OK and become reluctant to open their wallets and donate.

Then, the top Wikimedia Enterprise customers will acquire leverage over Wikimedia and try to get Wikipedia curated to their convenience. Wikipedia articles will start being indistinguishable from advertisement.

Governments will intervene and want their share of influence too.

Top volunteers will start asking to be paid, many others will leave, some others will become critics of the project.

People will start being skeptical of Wikipedia because of their biased editorial line and then the project will be declared a failure, once everyone is angry and a beautiful project is torn apart by greed.


The Foundation has already considered this. https://meta.wikimedia.org/wiki/Wikimedia_Enterprise/FAQ#How...



Accounting tables often do this. This is not the perfect example, but here's a flavor of that. (last page of PDF) https://s23.q4cdn.com/574569502/files/doc_financials/2021/q4...


Yep, understandable. Right now it kicks out pdfs that don't fit the rules, but I think there are a few sensitivity variables / configs I can incorporate to make that seamless.


Timing is everything :) Would appreciate feedback!


Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: