2,438 karma · joined May 5, 2021
The AI companies focused most their effort on writing software and continue to do so. Software, SaaS, and software engineers are the first to be disrupted.
> but AI has completely changed that.
Exactly. However, it is because AI is focused on solving writing software first which is the step to solving everything else.
Willing to relocate: Yes
Technologies: TypeScript, JavaScript, Python, PHP, React, React Native, Svelte, Express, Bun, Node.js, Tailwind, Angular, GraphQL, D3.js, visx, Backbone, jQuery, LangChain, Mastra, FastAPI, pandas, scikit-learn, Optuna, Chrome Extension API, Playwright, Electron, Stagehand, browser-use, Web Audio API, WebRTC, WebSockets, PostgreSQL, TimescaleDB, MySQL, MongoDB, Redis, AWS, EC2, S3, Lambda, Docker, Git, LLM agent design, agent evaluation, reinforcement learning, browser automation, MCP
Résumé/CV: Ask via email
Email: [HN username]@gmail.com
I'm looking for a competitive, fast-moving team to join. I'm also open to contract work, but I would prefer to commit full time.
If your approach to making decisions is to spend 30 minutes using coding agents to spin up several infrastructure configurations on AWS, then use automation to stress test them for 5 minutes to find the best configuration for running your system; or to set up a grid of front-end frameworks like React and Svelte, implementing every state management system, like Redux and RxJS, in isolation to build a performance table so you can make informed decisions with empirical data, then I would be a good fit for your team.
Portfolio of data visualizations:
| https://adamsohn.com/algoviz/
| https://adamsohn.com/grammar/
| https://adamsohn.com/reasoning-grid/
As a consultant and full-time engineer, I’ve led 0-to-1 product development across streaming, real estate, edtech, marketing, and media. Having worked at companies ranging from a 130-person AI organization to a scrappy 7-person team, I thrive most in fast-paced, high-ownership environments.
I've been doing a lot of traveling for the past 3 years and I agree with you that it is `definitely not "most"`. Everywhere people interacting with each other in third spaces and cafes even if it is street food. There are open air markets around the world filled everyday with groups of people interacting with each other. There are parks and squares around the world filled with families sitting together on benches. Around the world there are churches, mosques, and temples filled with families and groups of people.
Most places in the world people will happily make small talk or have a discussion with a stranger even if I only know 100 words of their language.
Sure many or most people in some places are working 10 hours a day 6 days a week but I don't think it is the isolation I see in the United States. It is the same in De'Nang Vietnam or a market in Lima, Peru when I went everyday to get coffee in the morning; it was always the same person working ,any time or day of the week, but they were always friendly and welcoming and interacting with with the same people.
Neil deGrasse Tyson used the number 8 as an analogy to explain that mathematical similarities between two systems do not mean they share a physical connection -- they were discussing how both the universe and human brain can be described using fractal mathematics. He pointed out that you can count 8 planets in the solar system and 8 children in a room, but having the same count doesn't mean the children are planets.
The similarities say more about mathematics than it does about the universe and the human brain or planets and children.
Will the United States suffer the same problems as the Soviet Union when ~15 people control and own all the ideas?
This might not be communism or socialism -- this is whatever the Soviet Union was. There is owning land which is tangible and easy to understand. Everyone owns the land together because it is the means to produce wheat. Owning ideas is hard to grasp. If a person thinks something unique and writes that thought on paper, they own that thought. Every single machine is an idea: the light bulb, the transistor, the little piece of plastic that holds the two ends of hula hoop together, and the blue LCD. Owning the idea of the process has an extra step, but it is still ownership of an idea.
> To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;
Owning our ideas and their derivatives is so foundational to the American identity, some Americans might say, "I ain't no communist!" In order to promote intellectual discovery, the Founding Fathers decided that people are allowed to own their ideas, the rights to use those ideas, sell them, and to profit from them.
For better or worse, ignoring the United States Constitution and ignoring the fundamental tenet of ownership of ideas is a non-violent communist revolution.
The clause is what ensures profits from market sales or licensing of ideas go to the creator.
Removing (or ignoring in the case of AI companies) that clause in the US Constitution is what abolishes private ownership.
The first time a saw a documentary about Tetris it really hit me what communism is -- nobody owned anything they invented or created. [0] It was a long time ago and I remember feeling sad watching the story. In the Soviet Union, a group of ~15 people, Politburo, controlled everything including any thought written to paper.
It is this one line, Article 1 Section 8 Clause 8, that separates the United States from the disaster that was the Soviet Union:
> To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;
I don't think it is far fetched to call ignoring and disregarding the Copyright Clause a communist revolution, violent or not. That is the one thing the communists -- there have been many over the years inside the United States -- would change to make the United States a communist country.
[0] https://en.wikipedia.org/wiki/Tetris#Spread_beyond_the_Sovie...
They have a theory that artists who have to struggle more during the training and developing of being able to express on medium develop into better artists.
> 512 shards, each with one Postgres primary each on an r8g.16xlarge
> 480 Neki routers, each on its own 8xlarge instance
> We sustained 118,538,803 QPS for 16 minutes across 512 shards and 1.22 PiB of data. Our largest recording was 118,747,267.
Component Detail Monthly Hourly 16-min burst
---------------------------------------------------------------------------------------------
Shard compute 512x r8g.16xlarge $1.41M $1,930 $515
Router compute 480x r8g.8xlarge* $661K $905 $241
Storage (gp3 floor) 1.22 PiB @ $0.08/GB-mo $102K $140 $37
Storage (io2 floor) 1.22 PiB @ $0.125/GB-mo $160K $219 $58
IOPS (io2, light) 5K IOPS/shard, tiered rate $166K $228 $61
IOPS (io2, medium) 20K IOPS/shard, tiered rate $666K $912 $243
IOPS (io2, worst-case) 231,517 IOPS/shard (0% cache) $4.56M $6,251 $1,667
---------------------------------------------------------------------------------------------
Total (gp3 floor) $2.17M $2,975 $793
Total (io2 floor) $2.23M $3,054 $814
Total (io2 + light IOPS) $2.40M $3,282 $875
Total (io2 + medium IOPS) $2.90M $3,966 $1,058
Total (io2 + worst-case IOPS) $6.79M $9,305 $2,481Florida politics? [0]
[0] https://civileats.com/2019/06/25/toxic-red-tide-is-back-in-f...
Remote: Yes
Willing to relocate: Yes
Technologies: TypeScript, JavaScript, Python, PHP, React, React Native, Svelte, Express, Bun, Node.js, Tailwind, Angular, GraphQL, D3.js, visx, Backbone, jQuery, LangChain, Mastra, FastAPI, pandas, scikit-learn, Optuna, Chrome Extension API, Playwright, Electron, Stagehand, browser-use, Web Audio API, WebRTC, WebSockets, PostgreSQL, TimescaleDB, MySQL, MongoDB, Redis, AWS, EC2, S3, Lambda, Docker, Git, LLM agent design, agent evaluation, reinforcement learning, browser automation, MCP
Résumé/CV: Ask via email
Email: [HN username]@gmail.com
Portfolio of data visualizations:
| https://adamsohn.com/grammar/
| https://adamsohn.com/reasoning-grid/
| https://adamsohn.com/separate/
| https://adamsohn.com/algoviz/
I bring 13 years of full-stack UI development experience alongside deep expertise in browser automation and agents, which I've been engineering since 2018. This blend makes me particularly strong in QA automation engineering and complex frontend architecture.
Notable projects include event ticket inventory management and a drag-and-drop CRM email builder for social marketing campaigns. As a consultant and full-time engineer, I’ve led 0-to-1 product development across streaming, real estate, edtech, marketing, and media. Having worked at companies ranging from a 130-person AI organization to a scrappy 7-person team, I thrive most in fast-paced, high-ownership environments.
I find once there is one poisoned line of text it negatively affects everything else downstream. Instead, I use a temp/ folder with documents and use different files for different agents and models. Then I have to constantly prune and delete the files. Any information that can be extrapolated is just noise which negatively affects the agent. If you have a definition of a database structure and it has been implemented, that information should not be contained in any text document -- it is noise, will drift, and be impossible to debug why the agent keeps producing undesired behavior.
I have a ~/Projects folder. For example, I use Playwright with Chrome DevTools Protocol in order to do performance testing and leak detection. There is a script that handles this. My prompt is "Search ~/Projects for perf testing with CDP and Playwright and implement here". Point being, if I need anything I point to a resource or ask to search a resource and it will find it quick and, most importantly, tends to improve it every iteration.
If I was in an institution, I would have a repository and would rather just point the resource and say use that than have memory of it locally.
Can you discuss this? I might be able to help.
It is also a very easy win for a company to just focus on the one stolen element or component in a lawsuit to knock competition out of the space and has happened multiple times.
[0]https://en.wikipedia.org/wiki/The_Short_Reign_of_Pippin_IV
I put it at the top of CLAUDE.md. I wonder if I put at a 8th grade level, it would be less of a cognitive load.
Humans are not born being able to achieve that. It is learned behavior. You and everyone else will remember their teacher saying, "Check your work!" Both the human child and the model work through multiplication problems using the same technique, using the distributive property. They both try to get a reward. For the human child, it is a sense of someone commending them for correctly solving the problem, a reward that probably yields some type of positive dopamine or serotonin feedback loop.
The model solving the problem will have a lower error rate if the first series of tokens created is followed by a series of validation tokens that are subsequently followed by error-correction tokens if there is an error!!!
Maybe it is thinking. Maybe it is remembering to validate and check the work and then remembering to fix the error. For the model trained with RL, why did tokens associated with validation towards the middle of a stream of tokens yield much better results? DeepSeek proved with R1-Zero that a model will learn to verify and correct itself from RL alone with no supervised fine tuning (SFT) teacher ever showing it how. The only reason DeepSeek used SFT was to clean up the reasoning tokens to be human readable. [0] When constrained by SFT, the models will use the double meaning of words -- polysemy -- to satisfy being human-readable while also carrying meaning for what they are working on.
Different people think differently. I watched a viral video of some ~11-year-old child talking to his mom or dad about a stream of a voice in his head. He discovered for the first time that he has a stream of thought. When he goes to school and solves a long multiplication problem, like the stream of tokens from the model, that voice will say to itself (him), "Check your work!"
That is a case of the stream of thought as words being aware of the stream of thoughts as words. Self awareness is a different conversation.
What I think is happening is that the child's stream of thought while solving a multiplication problem in school is likely very similar to an AI model's stream of tokens solving a multiplication problem. And they both were learned. The mechanics are very different, yet, the analogy is apt.
[0] https://huggingface.co/chutesai/DeepSeek-R1-NextN/blob/main/...
I created a flame graph classification of thinking-token phrases into setup, execution, decomposition, verification, error correction, surrender, and deliberation, or classified as steps in an OODA loop, which is more of a reach. It literally has a verification step and, if it finds an error, an error-correction step.
If there is a verification sequence of tokens with an error-correction sequence of tokens during RL training, it will perform better; and if humans do these steps (did you proofread your reply to this comment? did you correct it?), they will perform better — which is why it is so easy to make the anthropomorphizing metaphor.
Nonetheless, the paper is 100% correct that these machines are not thinking like humans.
In a fraction of time it takes me to solve any 20 - 40 line code problem, a coding agent can solve it 10 different ways in python and in TypeScript, inject performance logging, run each in 1,000,000 iterations with as many permutations as inputs, write comments at a 10th grade reading level so I understand what each does quickly, make a clean table with pros / cons and performance results, and I after considering the options choose one.
The problem is that the coding agents are not dependable -- they are reliably incorrect.
In the United States decades ago, a phone utility company was sued because they didn't allow women to be linemen working in the field. They lost and what they did was make changes like using lighter aluminum ladders getting rid of the heavy wooden ones, they replaced the wrenches with ones with longer handles so they had much more leverage, and many other things to make the work less physically punishing. A reporter asked some of the veteran linemen how they feel about working with the changes. The reply was, "why didn't we make these changes sooner?" None of them lost their job and their job got a whole lot easier.
The problem is verifying code quality. The coding agents can't reliably do it. But they as tools, can help both juniors and seniors make their job a whole lot easier.
Yesterday claude code built a console that steps through algorithms: one shot. There was a bug with a value being incorrect. I thought this would be a great way to automated visualizing and stepping through code during a PR review.
I'm sitting in a room with a computer by myself where I was thinking yesterday about a way that AI can add value to junior engineers. I see a post and discussion about junior engineer's value so I shared what I'm think and working on.
Hopefully I'm contributing to the conversation here and I can get feedback good or bad about how to approach improving junior engineer's value.
I developed a system to help prepare for leet coding interviews so I never feel lost under pressure solving a problem again. It is like a debugger that steps through the code showing all the values of all the variables with data visualizations that reflect the logic so I can grok what it is doing. [0]
After I had the Claude build it, I started looking at the values and there were some mistakes. So, again, the coding agent ran all the code, recorded all the values, and made sure that they line up.
Here is the really cool thing about that. The coding agents can't be trusted. By observing the values stepping though, what I really was doing was debugging coding agent code. It is debugging code presented in a way that is extremely simplified.
What I've been thinking about yesterday and today is, can I do the same thing with a pull request? Have the coding agent run the code, capture all the values, and create a console for the reviewer to step through looking at with data visualizations that abstractly represent that code.
Two things. 1. Coding agents can't be trusted and 2. reviewing code is very difficult. But is it possible to use coding agents to make reviewing code easy for humans? I think so.
That would be a great way for junior engineers to be extremely useful. They only have to step through the code and make sure that all the values line up.
So is nuclear annihilation. Yet, here we are.
> "I occasionally think how quickly our differences worldwide would vanish if we were facing an alien threat from outside this world." -- Ronald Reagan at the United Nations [0]
We are going to be fine as long as we remember to be nice to each other, starting towards the people who are adjacent to us.
I wanted to demonstrate capacity (how well it does a thing) instead of capability (which things it does, like drawing a pelican on a bicycle with SVG or solving a Rubik's Cube). To understand how LLMs solve math, look at the simplest case of multiplication. I deconstructed and classified the thinking token output. It is very important that model training yields thinking token output that structurally follows an observe, orient, decide, act (do the multiplication), and observe again loop.
A probability grid of chain-of-thought, read through Boyd's OODA loop lens [0]