Jay Gengelbach

The best use case for overseas development I've heard: We have customers in the eastern hemisphere. To support those customers better, we're standing up product, dev, and support teams to help serve those customers needs. Two key ingredients in making that successful: commensurate pay and the same level of performance and accountability as US peers. #software #international #teammanagement #overseasdevelopment #startup #leadership #coaching

6

I just realized: we should start saying "build memoization" rather than "build caching." We aren't really talking about least-recently-used or time-to-live algorithms, we're talking about "don't bother recompiling this piece if none of its inputs changed." (It has been puzzling me that some people are so deeply fearful of "build cache" as a concept, and honestly I think it's mostly because most build systems have non-hermetic state scattered all over the place. Maybe a slightly more precise term for the type of caching we're concerned with would clarify the semantics in question.)

83

9 Comments

Anthropic's latest Claude AI model is taking human-AI interaction to a whole new level! 🖥️🤖 The new Claude 3.5 Sonnet can now use computers just like a human would - by looking at the screen, moving the cursor, clicking buttons, and typing text. This is a major leap forward in AI capabilities. Key features: Operates based on screenshots, emulating human computer use Can perform complex multi-step tasks across different applications Self-corrects and retries when encountering obstacles2 Accessible via API for developers to build with Why it's game-changing: This opens up incredible possibilities for AI assistance in data entry, form filling, software testing, and automating repetitive computer tasks13. Imagine having an AI assistant that can seamlessly work alongside you, handling tedious computer work while you focus on higher-level tasks. While still in beta, this technology demonstrates how AI is rapidly advancing toward more human-like interaction with digital interfaces. As it develops further, it could revolutionize how we work with computers and boost productivity across industries. The future of human-AI collaboration just got a lot more exciting! What do you think about this new capability? #AIInnovation #FutureOfWork Here is a video about it: https://lnkd.in/gh8aNYEu

6

Its about solving the problem, not 'having AI' in the system. This is from last March, but the Wave 1 vs. Wave 2 distinction does seem to be playing out. Less 'generate me content', more 'synthesize vast amounts of relevant data and aid in decisonmaking or signal extraction'. https://lnkd.in/ehFBFC32 They don't talk about multi-agent systems here, but looking forward to more startups with a focus on leveraging agentic design patterns (https://lnkd.in/ed5rMmFe) to streamline real world workflows.

4

1 Comment

This product was practically built for me. I’ve been waiting for something like it for years, long after I gave up writing on paper almost entirely. Heck, if it wasn’t for having kids, I probably could have gone the last the past decade with the challenge of not writing anything down on paper. Which makes it all the more frustrating at the price point it’s released. This isn’t targeted for grade school students, college students, or young professionals. It’s targeted at prosumers which share the same space as an iPad Air, and the market is already dominated by prosumers who want that product. For those who can’t afford the iPad, Amazon releases tablet computers in the $100 range. I know Remarkable is a much smaller company, and can’t compete on price with Amazon, or volume like Apple. But waiting so long for something I can’t justify the purchase price just hurts. At half the price it would have been a no brainer. It checks practically every box.

11

5 Comments

Clear and rich information on RAG variations. I think this will be useful to most practitioners who have gone beyond the intro and demo workflows, and now need to consider how to refine their flows.

10

One potentially really good use for AI (for the rich person who wants to contribute back to US society): a model which asks for someone's opinions on general topics, and then correlates that with all available data for election choices, and fills out a sample ballot for the person based on the data. Each item could have a "drill down" link, to explain the available trade-offs, and why the option was selected. Rationale: A lot of voting is inefficient (in terms of "selecting the best option per people's desired outcomes"), because voters are bad at correlating large amounts of data, and propaganda people are good at confusing people and misrepresenting causal actions. But ML models are good at the former, and if trained properly, effectively immune to the latter. My assumption is that most voters/people are generally well-meaning, and trying to do the "right thing" for the country, but that's increasingly difficult in today's world of disinformation and propaganda. An unbiased model which aligned suggested choices with desired outcomes and trade-offs, based on actual data, would be potentially tremendously beneficial for getting better outcomes from elections in general. Note: This would not be an easy problem to solve by any stretch... but it could be a really positive change, imho.

1

One of the incredible topics I've been able to learn about, and teach, at Amazon has been implicit communication - the meaning that HOW you write and say things imparts to WHAT you write and say. Along the way, I've found quite a lot of information in the form of research studies, scholarly articles, books, factoids, tips, tricks, and so forth - like this article about how being facially expressive has been shown to make people more likeable and socially successful. However, the most valuable lesson I've learned isn't any of those tips, tricks, or facts. It's the importance of being empathetic and authentic: to understand my own emotions and adjust my reactions in response to others, and to instill in myself a genuine desire to collaborate that aligns my implicit and explicit communication. So much of implicit communication is hard-wired into us as humans and happens subconsciously in ways we can't control - but we can control our own emotional state and awareness of others' needs. What emotional state do you want to have going into any conversation or situation? How can you evoke that emotional state in yourself? How can you identify when you may have misinterpreted the communication of others? What can you do to address and mitigate the impact of that misinterpretation? #amazon #leadership #emotionalintelligence #communication #collaboration

18

2 Comments

"Jeffing" is a term that refers to the run/walk coaching technique founded by Jeff Galloway in 1973. Galloway is an Olympian and competed as part of the 1972 US Olympic Team in the 10km event. My blog this week explores how to get inspired by this technique to enhance leadership and management skills. Pace your team, encourage breaks and flexibility, and mix up tasks like alternating between hard running and easier walking. Set achievable goals, be flexible, and create an energized and healthy work environment for long-term success. https://lnkd.in/gWjR4x9U #leadership #management

49

6 Comments

As someone who has been working as a Professional Software Engineer in various roles for almost 4 decades, I no longer raise an eyebrow when somebody claims that some hot new technology + methodology is going to fundamentally shift the risks vs. rewards of designing, building, and deploying software services. But I have to say that the shameless hype-fest from OpenAI these last few months has been unrelenting, and now I'm seeing posts from otherwise reasonable-seeming folks about "PhD level" intelligence from GPT-5, like this one on June 20th: https://lnkd.in/eGkJDPGx And I'm tired of explaining why LLMs (although very useful for certain applications) aren't actually intelligent at all, because I'm never going to be able to spend as many hours breaking down the flood of blatantly self-serving + misleading statements being generated by LLM-based service vendors as they will be able to spend generating them. I should also add that LLMs are themselves the perfect tool for generating hype about LLMs, which I don't doubt is inflating the volume of nonsense to which we are all being subjected! So I propose the following "Questions for a Superintelligent Mind" as a way to cut through the GPT-5 hyperbole; IMO any entity which can actually give _correct + testable_ answers to (any of) these in a form that is comprehensible to humans deserves to be described as having both "PhD level intelligence" _and_ "superintelligence", since to date no PhDs have been able to answer (any of) them: 1. What are whales talking about? AFAIK there are plenty of recordings of whale song that can be correlated with information about the age, position, and health of the individuals who made the recorded vocalizations. But if you (superintelligent being, you) need more data, then design the experiments required to get it, write a proposal to the NSF to fund the field work, review applications from candidates to do it, analyze the results, and repeat as needed (you know, _do_ the science). 2. What is junk DNA for? Only 2% of the human genome corresponds to genes which are expressed via known transcription (e.g. to make proteins) and we have no idea what the other 98% does, except that mutations in certain parts of the "unknown" portion are associated with inflammation disorders (like IBD), and certain cancers. Again, there's lots of data available, so please explain if + how gene expression is "regulated" by the genome, and what role "junk" DNA plays in this (or any other) biochemical function. 3. Is the Collatz conjecture true? https://lnkd.in/evuUceUE The answer to this question (and any math question posed to a superintelligence) should include a human-readable proof. 4. How does sonoluminescence work? https://lnkd.in/eDFRRv2v The answer should explain the mechanism which actually produces the photons. And if more data is required in order to figure this out, then go get it (as in 1 above).

13

4 Comments

After listening to Tanya Reilly's "The Staff Engineer's Path" (https://lnkd.in/g2VfTtCj), I learned that as an individual contributor who aspires to be a Staff Engineer, excellence in technical knowledge is insufficient; you must also have social skills to gain buy-in on every technical decision you make. It's also great to learn about concepts like "Social Capital" and "Influence at Scale" from there. Highly recommended for those who are inspired to level up their the individual contributor role.

12

My approach to interviews is atypical: I don't start with a fixed list of questions, but make questions up on the fly based on what would give me the most signal to make a good decision. It's inspired by Bayesian Optimization. More details in the article, thanks to Sam C. and Agnes Chong for the feedback. Summary: To get the most out of interviews, use the answers to your questions so far to guide your next question. Make your questions up on the fly. Ask yourself: “What is the question that would give me the most signal about whether this person would succeed in the role?” This approach requires practice and skill (and best applied after your first 100 interviews), but can help you get considerably more out of the interview than a fixed list of questions. https://lnkd.in/gBM5af93

157

20 Comments

You don’t normally see LLM researchers and engineers talking about their day-to-day work. Big shoutout to this YouTube video from the Anthropic LLM Interpretability team, the team behind the Golden Gate Claude. It’s amazing to see that most of the team members have worked for the team for only one year, and yet they have made such a big breakthrough in interpretability research. Another thing I want to highlight is the unique challenges facing the team and how they approach these challenges, that is, through fast trial-and-error iterations and constant evaluation of what may work and what to prioritize next. #artificialintelligence #machinelearning #deeplearning https://lnkd.in/eUaPAYch

98

1 Comment

I'm excited to share my latest article on LeadDev: "From False Urgency to Strategic Leadership." In this piece, I delve into the challenges and opportunities that come with shifting from reactive problem-solving to proactive strategic thinking. This article is designed to be a resource for staff+ engineers looking to enhance their leadership skills and drive meaningful impact in their role. 🔗 Read the full article here: https://lnkd.in/g7qxqE9r I hope you find it helpful and I’m eager to hear your thoughts and experiences. #Leadership #TechLeadership #LeadDev #FalseUrgency #StrategicLeadership

50

3 Comments

Honest one liner thoughts for leaders and individual contributors in data: 1) Under-documented knowledge is toxic and can kill otherwise solid teams. 2) Consider the overall cost AND impact of your tech stack on the business' bottom-line, thinking in terms of tech, labor, and opportunity cost. 3) Best practices are awesome and can bring benefits down the line, but when running a business, the right answer is the solution that just gets the job done fast. 4) Beware of vendors that tell you worksheets will be gone forever if you implement their solution. Love to hear any other thoughts!

18

6 Comments

Developers can use this feature to designe mockup and implement some basic features.

1

“If microservices are a well-conducted orchestra, Agentic Architecture is more like a jazz ensemble, where each musician not only plays their part but also listens, adapts, and improvises based on what others are doing. In an agentic system, the components (agents) are autonomous entities capable of making decisions and taking actions on their own.” New article about Agentic Architectures: https://lnkd.in/dQ_jgSyg Subscribe to www.zero-bits.org and get the articles as they are published. #agentic #architecture #ai #software

20

The Domino Effect of a Keystone Habit "It wasn’t the trip to Cairo that had caused the shift, scientists were convinced, or the divorce or desert trek. It was that Lisa had focused on changing just one habit—smoking—at first. Everyone in the study had gone through a similar process. By focusing on one pattern—what is known as a “keystone habit”—Lisa had taught herself how to reprogram the other routines in her life, as well. ...'All our life, so far as it has definite form, is but a mass of habits,' William James wrote in 1892. Most of the choices we make each day may feel like the products of well-considered decision making, but they’re not. They’re habits. And though each habit means relatively little on its own, over time, the meals we order, what we say to our kids each night, whether we save or spend, how often we exercise, the empathy we automatically show others, and the way we organize our thoughts and work routines have enormous impacts on our health, productivity, financial security, and happiness. One paper published by a Duke University researcher in 2006 found that more than 40 percent of the actions people performed each day weren’t actual decisions, but habits." ― Charles Duhigg, The Power of Habit: Why We Do What We Do in Life and Business https://lnkd.in/dAvs5PiD #habits #continuousimprovement #growth

8

7 Comments

Something that only became obvious to me after I had been writing software for a couple of years is that there are different types of bugs, with different characteristics, impacts, and mitigation strategies. One class of bug that doesn't often get talked about is what I call the Semantic Bug - a Semantic Bug is basically an "oops I made some assumptions about this data/API/module that aren't really true" bug. For example: - "oops I thought this method was idempotent but it's not" - "oops I didn't think users could be active in two teams at once but they can" -"oops I didn't think these events could happen in that order but they did" I find it helpful to think of a dichotomy between logic bugs ("the code doesn't do what I intended") and Semantic Bugs ("the code does what I intended, but in the context of the larger system the thing I intended is wrong"). You might not run into Semantic Bugs too often when, say, working on a personal project, but they can be a real problem when developing larger software systems - especially when you have to reason about models and interfaces owned by other teams. Some types of bugs are fairly benign - if some unicode formatting looks weird in production, for example, you should fix it, but it's not going to break any functionality. It's generally ok to just squash these bugs when they show up and move on. You don't want to let that happen with Semantic Bugs, though, because they can be pretty nasty and unpredictable: -They don't always lead to a simple crash. In the worst cases, they can cause faulty components to exhibit essentially undefined behavior, potentially lying to other components about the state of the world in a way that’s hard to detect. Bad behavior can spread through the system, causing it to corrupt data or take inappropriate action. -They can be hard to catch with tests, if only for the reason that you’re apt to make the same faulty assumptions writing your test cases that you made writing your code. -They’re pernicious. Sometimes your wrong assumptions may only be wrong under very specific conditions, and you may be (un)lucky enough that it takes months for those conditions to occur in production. This can be maddening - when bugs occur directly after deploying a code change, you can at least point to a probable culprit, but when these “sleeper” bugs hit it feels like your happy, stable system keels over and dies for no reason. Semantic Bugs are really a product of our fallible human brains more than any technological constraint, so it might seem like there's not much we can do to prevent them besides making sure we're communicating well with our partner teams. The good news is that there are, in fact, architectural techniques you can use to make your software naturally resistant to Semantic Bugs! We put up a survey of existing approaches (along with a couple new ideas) at https://lnkd.in/g4mjqw2a - check it out if you're working on a large codebase!

14