Dan Moldovan

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

Struggling to manage Kafka configuration in large engineering ecosystems/environments? Things are breaking up, and you can't roll back or save the situation quickly? Our own Software Architect, Nizan Rosh, just published an article in the #LSports engineering blog discussing how we solved these problems with a super simple yet elegant #GitOps tool. We also published it as an #opensource to top it up with a cherry!

35

2 Comments

Excited to share how I've optimized my workflow with my home lab setup! 🚀 My CI/CD tooling, along with automations for Jira and Google Docs integration, has been a game-changer. While Jira is fantastic for workflow management, JQL language can be limiting. But fear not! With JQL, you can extract any subset of Jira data into Google Sheets seamlessly with simple queries. DevOps moving to DevEx, making work more fun every day! :) Check out this insightful article on Jira Smart Chips in Google Docs for more details: https://lnkd.in/e9gDePRt #WorkflowOptimization #JiraIntegration #CI/CD #Automation #GoogleDocs #TechSolutions #DevEx

11

There are a lot of hyped up claims about AI being tossed around lately; especially when it comes to software engineering. After being deeply immersed in this misunderstood industry for over 20 years, I can say whole heartedly, don’t believe the hype. This article does an excellent job at describing what the craft of developing software is really like and why the reports of its death have been greatly exaggerated. https://lnkd.in/eY-59vxH

14

Announcing the new #SpatialAI capabilities in #AutonomousDatabase! Join David Lapp in our #ADBLearningLounge session next week: https://lnkd.in/eKKYNu9Q We will show new specialized #ML #algorithms that incorporate the effects of location, improving ML model quality and prediction accuracy. An example of spatial regression algorithms that are able to enhance home value predictions by incorporating the influence of neighboring home values will be shown. Don't miss this opportunity! #oracle #adb #spatial #ai #learninglounge

13

1 Comment

We released a new update of our Visual Studio Code Java extension. Go try it! Bunch of new improvements and already supporting JDK 23 Early Access builds as the extension is directly leveraging javac. https://lnkd.in/g8BKsNjS #java #IDE #OpenJDK #VSCode

61

1 Comment

This needs to be said. Investors and CEOs have failed to learn from history. They're repeating stupid economic mistakes from earlier decades, and we are all suffering (or will soon suffer) the fallout. The fundamental lesson not learned is this: value is only created by talented humans, and that talent is inherently expensive. If you refuse to pay for the talent you need, reality will catch up with you and smack you down hard. A.I. is not nearly as capable as investors and leaders think. It is not going to replace human labor costs. It cannot reason, solve problems, or write code to solve new problems. It cannot make correct judgment calls. It cannot invent or innovate. Boosting profits by cutting human labor costs is a temporary and self-injurious hack. Products and services do not magically keep improving or evolving without the talented people who make and operate them. Customers do not want to pay more for, or buy more of, offerings that are declining in quality and meeting their needs less. You must give them a compelling reason to keep paying you. Quality plummets when you cut R&D or QA entirely or outsource them to the lowest-cost contractors or remote teams you can find. "Never go with the lowest bidder" is sage advice for good reason. Profits are a long-lagging indicator of the investments they are a return on. Cutting investments now may make the next couple quarters rosier, but that lack of investment will come back within 1-2 years to hammer your business in the worst ways possible, and by that time it might be too late to recover. None of this is unpredictable. None of it is a pattern we haven't seen play out before. None of it is new. We all know where it leads: severe economic recession, reduced GDP, fewer jobs, and a poorer quality of life for everyone. Instead of cutting costs, boost your revenue. If your profit is falling, flat, or "not growing fast enough", recognize it's due to the enshitification of your products and services rather than to rising human labor costs. Fix the term in the equation that leads to growth; don't focus on the one that can only ever be a race to the bottom! Halt, or even reverse, the enshitifcation of your offerings -- by making the correct *increased* investments now and letting people who *understand the customers* call the shots -- in order to boost revenue and profits in the future. Be smart and make moves that produce value rather than destroy it.

12

🌟 Attention Engineering Managers and Technical Leaders! 🌟 In just two weeks, on June 28th at 10 AM PDT, I’m excited to host an interactive roundtable on Feature Flags and Delivery in conjunction with ELC! 🚀 Why Attend? This is a collaborative discussion where your expertise and curiosity are key! Whether you're experienced or eager to learn, your contributions are invaluable. 🔍 Discussion Points: - Decoupling Processes - Boosting Team Efficiency - Optimizing Feature Delivery - Tracking Platform Usage Expect rich conversations on best practices, unique implementations, methodologies, and the latest tools enhancing feature flagging and delivery. Who Should Join? - Engineering Managers - Technical Leads - DevOps Engineers - Software Developers This roundtable is your chance to: - Share your experiences - Gain new insights - Network with peers - Discover practical recommendations RSVP Now and mark your calendars for June 28th at 10 AM PDT! Let’s collaborate and elevate our skills together! #EngineeringManagement #TechLeaders #FeatureFlags #DevOps #SoftwareDevelopment #TeamEfficiency #TechRoundtable

9

I just blogged about the benefits of contract-first API development. If you and your team aren't following this approach, I encourage you to read on and learn why you may be missing out on a practice that can help your team be more productive.

6

Abstractions leak in the worst case. So it's always good the low level details behind the abstraction. Despite this, abstractions are super important: What if there was no abstraction in computer networks. Imagine every backend developer implementing TCP in their web server. It would be total chaos!

8

2 Comments

Current conversational voice assistants provide high quality interactivity and feel much like a real person. Large Language Models (LLM) can supplement a conversation to "think" through and reflect upon those conversations. In this video I have a conversation with my Synthflow AI assistant. We chat about a business idea and then the assistant sends the call transcript to Zapier. The Zapier workflow uses some tools, especially Anthropic's Claude, to create a more formal business plan from the conversation and sends the PDF to my email address. A voice assistant's ability to converse in conjunction with an LLM assistant's ability to understand are a formidable team. Using these tools we can consider a business idea quickly and more fully than we have in the past. https://lnkd.in/gyFU3pSn

15

2 Comments

How to live a better life for yourself and others

1

I have a new blog post that discusses our journey in speeding up Docker builds for a large rails monolith - https://lnkd.in/gT9HVnWv

18

Project update! My application tracking extension is now doing fuzzier matching and injecting new information into LinkedIn (and other sites) to keep me from doubling up on applications. Next up: local storage and bulk import/export operations...

12

2 Comments

Our own Denis Leconte shares some AI-infused advances in digital archiving. https://lnkd.in/gFV_PW6K #artificialintelligence #ai #technology #archives #digitalarchives #smartvault #amics #automatedmediaimagecapturesystem #searchtools #leconte #denisleconte #ironmountain #mediaandarchiveservices #mas #ironmountainmediaandarchiveservices #post #postmagazine

16

Use-case for ChatGPT. https://lnkd.in/gJzgNbbV Am in the process of learning the nuts and bolts of "Using ChatGPT" and plan on sharing my findings here (and in other places).

1

1 Comment

A super clear, straight forward example of how to do cost effective RAG using LangChain for your document pipeline and Groq for compute resources. The code demonstrates how to speed up RAG and reduce costs using quantized embeddings (reducing 32 bit values to 8) Nice job Sahibpreet Singh! https://lnkd.in/gf5QMCKK

44

2 Comments

It is important to revisit concepts that might have eluded me. Take the Intuit Lacerte Software Development Kit (SDK). For the non-programmers, a Software Development Kit is instructions and tools for connecting to local data sources. Think of them as APIs but for accessing local data and hardware instead of cloud data. The first time I looked at the Lacerte SDK over a decade ago, I did not understand how it worked. The best I was able to do was to pull basic client information (names and address) for generating mailing labels. This was before my firm was doing digital delivery. As I started using Robot Process Automation (RPA) tools to input data into Lacerte, I learned about batch input. This let me gain a new understanding of the structure of Lacerte's data. Just last winter, I built some RPA tools to read organizer PDFs from Lacerte to build Document Requests List in SafeSend and Firm360. This let me gain a new understanding of the flow of Lacerte's data. Just a month ago, I connected with TruePrep AI to build tools for inputting data into Lacerte. My original plan was an RPA method since it was something I was familiar with. In that moment, I had decided to take another look at the Lacerte SDK. With an additional 10+ years of experience in understanding Lacerte's data structure and data flow, the SDK all makes sense. In my excitement of applying new knowledge, I can now generate those previous Document Request Lists more efficiently as well as add, update, and delete data from Lacerte. What does this mean? It means, custom import tools that can be catered to a firm's workflow like trial balances, depreciation, etc. It is always important to revisit concepts that have been previously elusive. We are the sum of our experiences and even a few months (from winter to spring) can change how we find solutions to problems.

31

3 Comments

Ever wonder how exactly your knowledge sources are being used during RAG? We did too. I'm excited to share our early work in explaining LLMs, "RAGE Against the Machine: Retrieval-Augmented LLM Explanations"! 🎸✨ 📄 Paper: https://lnkd.in/eV4qF8YV 📺 Video: https://lnkd.in/e6mNV8tt Last week at ICDE 2024, I introduced RAGE, an interactive tool designed to explain the outputs of LLMs augmented with retrieval capabilities (RAG). RAG needs explaining since it is unclear how the presence and order of knowledge sources (among other things) affects the LLM's answer! 🤖💡 RAGE helps users understand how LLMs generate answers by identifying parts of the input context that, when moved or removed, change the LLM's response. This counterfactual approach makes AI decision-making more transparent, and allows us to derive a form of citation. 🔄📊 Key features include: - Combination Tests: Which sources lead to which answers? 🔍 - Permutation Tests: How does the answer change if sources are reordered? 🔀 - Interactive Demo: Explore behaviors of real LLMs using real data. 📚 Whether it's determining the greatest tennis player or answering your own unique questions, RAGE supports various use cases. Our code (coming soon) is designed with interfaces, making it highly adaptable to any LLM, retriever, or text data source! 🎾🏆 For technical details and implications of our findings, please check out our paper. Along with my co-authors at University of Waterloo, York University and AT&T, we are working on several extensions to explain many unexplained aspects of LLMs, RAG, and beyond. Stay tuned for more updates and an open-source RAGE code release in the near future! 🚀💡 #ExplainableAI #LLM #RAG #Research

14

2 Comments