Kristoffer Gronowski

More and more of this will get into various markets and the burden of owning a car and having it parked kind of falls away. But how will the insurance markets deal with pricing this for any insured event. For example two driver less cars hit each other and create third party losses.

3

1 Comment

I’m skeptical about measuring open-source package quality with metrics alone.    •   Low commit/update cadence: Doesn’t indicate low quality. It might signal a stable, well-functioning package that needs minimal maintenance.    •   High number of historical CVEs: Not a sign of low quality. It could show that the package has undergone thorough security scrutiny.    •   Many open issues: Not a reliable indicator. Many issues are more like support tickets, and there are often many duplicates. Quality assessment requires more than just metrics!

3

One key component of Gearflow's success is our strategic decision to center the engineering team around product. This informs how we evolve the problem and solution space together while grounding the team in the proper motivation. We use a few key principles to drive innovation and distinguish ourselves from other product and engineering teams. Although our methods continue to evolve, the core idea remains: equipped and empowered teams allow you to do more with less people.

32

2 Comments

Confession: I am lazy. A while ago I showed a Prescient Designer/Node-RED subflow with multiple "virtual" inputs. Not too hard, but not simple enough for me. So I made a couple of nodes to make this super fast and dead simple. Watch how in my latest video! https://lnkd.in/e7WxhxpH #designer #nodered #lazy

14

2 Comments

"There is another element in this scientific attitude: only that is knowable which is expressed (or, at least, can be expressed) in numbers. To get away from the so-called 'arbitrary and subjective', to escape ethical or literary judgments (which, as everyone knows, are trivial and unfounded), the scientist must get back to numbers. What, after all, can one hope to deduce from the purely qualitative statement that the worker is fatigued? But when biochemistry makes it possible to measure fatigability numerically, it is at last possible to take account of the worker’s fatigue. Then there is hope of finding a solution. However, an entire realm of effects of technique—indeed, the largest is not reducible to numbers. Yet, since what can be said about it is apparently not to be taken seriously, it is better for the scientist to shut his eyes and regard it as a realm of pseudo-problems, or simply as nonexistent. The 'scientific' position frequently consists of denying the existence of whatever does not belong to current scientific method." — Jacques Ellul, The Technological Society یعنی وہ سمجھ سکے تو آنسو، نہ سمجھ سکے تو پانی۔

2

👩🏽‍🍳 We've been cooking! Embeddings are now supported in Tecton. This is just a first milestone, but an important step for Tecton bridging the gap between using structured and unstructured data in sophisticated applications using both predictive ML and generative AI. 🔢 Features represent structured contextual data used by models, whereas embeddings are powerful representations of unstructured, high-dimensional data. Productionizing embeddings at scale has significant technical challenges that we're actively solving to remove obstacles to building world-class AI products, starting with turnkey support for generating and serving embeddings in production. 🫡 to Brian H., Brian Seo, and Mihir Mathur and the team on this launch! https://lnkd.in/eugrpVdA #embeddings #features #models

69

4 Comments

This is really great and resonates - thanks Bryan Solar for the link! Couldnt align more to this - why add YAF (Yet Another Feature) when you can refine, prune, and streamline a feature in your existing product offering... On the engineering side this is also beneficial as it is less Cognitive Load and more time to cover Engineering PASS (Performance, Availability, Security, & Scalability)

15

I ran across this post in the Changelog newsletter: https://lnkd.in/eCdjVekq - I like the name slop for unwanted AI-generated content. I love LLMs. They are part of my daily workflow, and I've been exploring them more seriously while on my sabbatical. I wouldn't misrepresent generated content without critical review as my own, but the web seems to be filled with more generated garbage (or 'slop')—content that looks fine at first but isn't accurate or helpful. LLMs (and other generative AI) should help us to make things of higher quality, not just churn out slop more quickly.

11

1 Comment

We all say we're outcome-driven, user-centric, evidence-led, and laser-focused on solving real problems. Yet, these anti-patterns seem to creep in far too often: 1️⃣ Outputs over outcomes: Shipping features no one uses, but hey, velocity is up! 2️⃣ Opinions over evidence: Decisions made because someone really likes the idea (HiPPO effect, anyone?). 3️⃣ Prescriptive roadmaps: Rigid plans that crumble the moment reality kicks in. 4️⃣ Deadlines over goals: Delivering on time but missing the point. 5️⃣ Solutions over problems: Building shiny things with no clear problem to solve. What's the worst product management anti-pattern you've seen in the wild? #productmanagement

32

10 Comments

Unveiling Mistral AI Large 2: A Game-Changer in AI 🌟💼 Mistral Large 2, a groundbreaking AI model that challenges the biggest names in the industry. With just 123 billion parameters, it claims to match or even surpass the performance of models from OpenAI and Meta, thanks to its efficient design and advanced capabilities. Key Features and Takeaways: 💻🤯 💥 Despite having less than a third of the parameters of Meta's Llama 3.1 405B, Mistral Large 2 outperforms it in code generation and mathematics. 🌐 Offers impressive multilingual support for 12 languages, including French, Spanish, Arabic, and more, as well as 80 coding languages. 🔧 Enhanced with a 128,000 token context window, enabling it to understand and generate long-context responses. 📝 Mistral Large 2 minimizes "hallucinations" and provides concise, accurate responses, setting a new standard for AI communication. 🌟 Available for trial on Le Chat and accessible on major cloud platforms, offering a glimpse into its impressive capabilities. Why It Matters: The release of Mistral Large 2 is significant as it showcases the power of efficient model design. With benchmarks comparable to GPT-4 and other leading models, it proves that size isn't everything. This puts pressure on closed-AI leaders to innovate and improve their offerings. By making the model accessible on cloud platforms, Mistral AI is disrupting the industry and providing developers and businesses with a powerful tool. The efficiency and performance of Mistral Large 2 highlight the potential for more sustainable and accessible AI solutions, opening up opportunities for a wider range of applications. #mistral #ai #llm #aiinnovation #nlp #languagemodels #innovation https://lnkd.in/gTR3AmrB

1

Edit: Not my product, just something that I have seen! Interesting kick-started item. Unsurprisingly, I have a lot of aerodynamicists in my network and a few with specific interest in cycling. What do we think? I can see a drag reduction from this but I wouldn't know how to quantify it. How much energy will you save in chocolate bars over a mile? Is there a crosswind stability issue to be concerned about? #aerodynamics #cycling

5

21 Comments

We're not always collaborating when we think we are. This is interesting approach to the scale/model which describes three levels between which each collaboration moves. https://lnkd.in/dWPjfSsX

11

My favorite LLM-for-coding prompt: "write a unit test for this class." This is one of the highest bang-for-buck prompts I get out of Claude. I'd say about 80% of the time, the generated test doesn't actually pass--but it's still an excellent use of time. Although it's often incorrect about what values the asserts should be testing for (which calls into question its critical thinking; see my prior work), what it gets right is the shape of a test. It creates good mock objects that stub out behavior that's not critical to the test. It does all that boiler-platey stuff that sometimes makes test writing tedious. It gives you a starting point from which you can copy and paste test methods and add the correct asserts. The reason this pays such big dividends, though, is because the AI is only mediocre at writing code. Its autocomplete is powerful and a big time saver, but it often autocompletes something sensible and compile-able but wrong. Your trust level for AI-generated code should be low. But what is the smart engineer's solution to code we don't trust? Good tests. If I generate a set of tests I'm confident in, I don't have to read the implementation all that closely. If it has basically the structure I'm expecting and it passes my tests, I don't have to pore over the implementation looking for off-by-1 errors. At some level, I just trust: "If the implementation passes this set of tests, it's a good implementation." Smart engineers have long known that good tests speed us up. LLMs have not altered that reality in the least. And using LLMs to speed up the things that speed you up is a tremendous force multiplier.

142

20 Comments

Hot off the press, this latest Silicon Valley Product Group article is a good reminder that there are multiple paths to success, and we have to tailor our strategies to fit our unique circumstances. There’s rarely a single answer or approach when it comes to topics like organizational design— after all, people and companies are complicated and are never the same! To quote Marty: “I believe strongly that with the right people with the right skills, a strong product team can succeed no matter the org design. […] I will never again doubt what a motivated group of empowered and skilled cross-functional professionals can accomplish.” Definitely worth a read if you have ~4 minutes and lead or are part of a product team.

4

🔬🖥️ Google Cloud is teaming up with Bayer to revolutionize radiology through AI-enhanced tools, promising to significantly alleviate the workload for radiologists and improve patient care. This collaboration leverages Google's cloud computing prowess and Bayer's expertise in healthcare to develop a platform focused on radiology applications, such as CT scans, ultrasounds, and MRIs. Set to launch later this year in the U.S. and Europe, the platform will support developers in securely experimenting with and analyzing massive volumes of healthcare data, constituting about 90% of all healthcare data. It will also include features for AI-powered searches of scientific papers and regulatory documents, assisting in swiftly acquiring necessary regulatory approvals. https://lnkd.in/dUMeuguN #HealthcareInnovation #AIinHealthcare #Radiology

55

1 Comment

I'm an AI skeptic, yet here we are, navigating a world increasingly entwined with algorithms. There's a curious push to infuse AI into everything—product managers saying, "Let's add some AI," as if it's a pinch of spice for any dish. It makes me pause. Does every creation truly need that sprinkle? I find value in AI when it stands as itself—pure and purposeful. Like Jerry and Todd, distinct voices that contribute without overshadowing. AI feels akin to a rubber ducky. Not just a simple toy, but a companion that helps us think through complexities. It doesn't just hand us knowledge; it invites us to discover wisdom by uncovering connections we might otherwise overlook. Do we need an AI-powered chatbot on every corner of the digital landscape? Must every app and website greet us with a talking assistant, diverting us from the essence of our experience? Sometimes, I think, less speaks more. I remember when everything suddenly needed the internet—a connected toothbrush, of all things. It made me wonder: just because we can, does it mean we should? Perhaps with AI, we ought to tread thoughtfully. Embrace it where it truly enriches, but let’s not lose ourselves in adding layers without meaning.

4

1 Comment

🌐 There’s been growing concern lately about AI "training itself" into oblivion by feeding on AI-generated content, as if we’re on the verge of a digital "mad cow disease." But this fear feels exaggerated. Think about how we, as humans, learn: we are, in essence, trained by other humans who learned from yet other humans. This recursive nature of knowledge doesn’t lead to our mental degradation; instead, it helps refine and expand our understanding. We build on prior ideas, just like AI models do with their training data. The difference is that AI processes far more data, far faster, than we ever could. So, just as humans don’t lose coherence by learning from those who’ve been taught before them, AI won’t suddenly collapse into chaos because it trains on its own output. We have safeguards, just like human education systems have frameworks, curriculums, and critical thought to prevent echo chambers. In both cases, structure and checks keep things in line. Of course, it’s wise to remain cautious. Just like how language evolves over time, symbols and their meanings can shift. For instance, the tilde (~)—often overlooked as a simple, almost trivial character—has a much deeper history. It was originally devised in the late 19th century by Matilde von Rothschild to represent an intricate cross-stitch pattern. Isn't it funny how something that once stood for precision and complexity is now used as a symbol for approximation? But we shouldn’t let the Matilde Mark’s origin be forgotten. In the same way that AI needs to understand its lineage to avoid losing context, we should pay homage to the origins of the symbols we use. Because sometimes, the smallest marks on a page carry the richest histories. #AI #MatildeVonRothschild #HistoryOfSymbols #CrossStitching #TechEvolution

6

2 Comments

#ThrowbackThursdays Join us for the latest episode of Mob Mentality, where we dive deep into "The 3rd Habit of a Highly Effective Mobber: First Things First (7 Habits)." Our conversation explores the importance of scheduling your priorities instead of prioritizing what's on your schedule. We discuss the Eisenhower Matrix and how it can be used in mob programming, and we also examine the urgent and important trap. Additionally, we explore how to prevent stress, burnout, and constant fire fighting, as well as how to build deep and meaningful relationships. Throughout the episode, we provide practical tips for improving productivity and preventing crises and fires. Don't miss out on this insightful discussion. https://lnkd.in/dwUbqe_q Chris Lucian

1