Jacek Materna

Agents are getting really really good at programming, try this agent out: https://buff.ly/4cY7vRk I use this agent every single day multiple times a day to build functions with documentation, error handling, and type validation. It's overpowered.

1

The most downloaded datasets from the California HHS site? If you follow me on LinkedIn, you've probably guessed it - the Medi-Cal enrolled provider list and Suspended and Ineligible list are the top two, followed by other datasets by a wide margin. The California Health & Human Services Agency (Cal HHS) has a nice data hosting site that resembles data.cms.gov. The site looks slick and provides interesting insights, including visitor statistics and download metrics. It's fascinating to see how people are interested in provider eligibility and exclusion data for Medi-Cal billing. Looking at how organizations use these public datasets, compliance emerges as the dominant use case. This makes sense - when accessing government data portals, users are typically verifying facts related to government programs. The user base spans health plans, hospitals, IPAs, medical groups, and even commercial payers, as Medicaid suspensions often trigger reviews across the healthcare ecosystem. What's interesting is that this pattern extends beyond healthcare. On data.ca.gov, which hosts all kinds of government data, the most downloaded dataset is the directory of certified electricians. This shows a consistent theme - whether it's healthcare providers or electricians, people come to government sites primarily to verify professional credentials and eligibility. This raises an intriguing observation about how government data is used. The way organizations use these datasets - downloading lists to verify credentials and eligibility - is functionally similar to what modern Identity Provider (IdP) systems do, but through a much more manual process. What are your thoughts on the future of government data in healthcare identity verification? #Healthcare #DigitalTransformation #Compliance #HealthTech #GovTech #Identity

24

Have you reduced overhead by 99% on any meaningful back office task since the launch of AI? Well, why not? Apparently, Amazon has. The interesting thing about this new technology is that it shows the jobs most at risk are things like coding, where there is a clear, quantifiable feedback loop between action and result. But adoption is going to be the challenge. Honestly, a lot of small businesses don't even know how to do basic things like mail merge. Even my peers at large companies with better-than-average technical skills have barely used these AI tools. So far, I've reduced several back office tasks by 80% through the at least partial use of AI. It's amazing how powerful this technology can be.

1

🚨 New Financial Conduct Authority research reveals a concerning truth: #AI systems in finance may be more biased than we think, even after debiasing attempts. Key findings that matter for fintech leaders: • No single method captures all bias - multiple testing approaches needed • Traditional debiasing techniques can actually harm model performance • Word embeddings retain bias even after attempted neutralization • Customer-facing AI systems need rigorous bias testing before deployment Most striking? A chatbot could give different responses based solely on how customers phrase their queries. In financial services, where fairness is crucial, this isn't just a technical issue - it's a trust issue. 💭 What bias testing methods are you using in your AI systems? Share your experiences below. #AIinFinance #ResponsibleAI #FinancialServices #GenAI

16

2 Comments

Few agencies without a technical lead are able to get by (and none thrive). Relying on contractors can be a double-edged sword: they bring specialized skills but often lack a cohesive vision that aligns technology with your customer’s long-term goals. This disconnect results in software bloat, where unnecessary features and technical debt pile up. The pain is real: without a consistent, forward-thinking approach, your business is at risk of compounding technical debt that will stifle innovation. Don’t let short-term fixes lead to long-term problems. It’s time to rethink how you approach technology, ensuring it supports—not stifles—your growth.

15

2 Comments

If Texas passes this bill in January, it will only be the second state to regulate AI based on risk factors, which is how AI must be regulated without damaging the development of safe and responsible AI systems. I haven't read the legislation but the description in this article sounds better than most. California demonstrated that they can't be trusted to regulate high-risk AI like consumer LLMs, which is where almost all of the risk is found in AI. Unfortunately, Congress has thus far also failed in its most important responsibility with regard to AI, which is the safety and security of the American people, and also critically important -- failed to protect property rights and human rights. Antitrust regulators deserve credit for finally beginning to enforce laws. However, on the broader economic front, Congress continues to fail to provide an environment that promotes competition and innovation, and instead favor illegal behavior and dominance by oligopolies. I'm not suggesting that it's easy, particularly in terms of political pressures and financial influence from companies worth $trillions, but the U.S. WILL eventually collapse if we continue on this trajectory. As I've said many times, failing to stop consumer LLM chatbots immediately until they can be proven safe was the worst failure of governance related to technology and commercialization in U.S. history. The Biden admin and Congress share that legacy. Hoping that catastrophic risk doesn't manifest during a politician's term obviously isn't acceptable policy. Neither is unleashing among the most powerful, high-risk technology in history to the public, including adversaries and all evildoers, but that's precisely what Sam Altman did, and his punishment was to attract tens of $billions to OpenAI and be treated by Congress and the White House as a celebrity. It's not shocking then that many have followed. The incentives provided by the USG are beyond perverse -- they are literally suicidal in nature. Does that sound like behavior that inspires confidence?

2

3 Comments

U.S. digital bank Chime is launching a product that will allow customers to access up to $500 of their wages before payday, the latest move by a financial technology company to enter the fast-growing "earned wage access" market. The product, which will be rolled out in coming months, comes as Chime is preparing to become a public company in the "not too distant future," CEO Chris Britt said in an interview with Reuters. #chime #neobank #us #digitalbank #reuters #fintech #fintechnews #kostyuchenok

1

As software development grew in the 70s & 80s, it adopted management styles from manufacturing with little change. This gave rise to the Waterfall development model, emphasizing linear processes, extensive planning, and rigid workflows. These practices left a lasting legacy, particularly the focus on outputs over outcomes. In manufacturing, it’s easy to measure how many widgets are produced but harder to assess their impact on customers. Applying this output-focused mindset to product development is a critical flaw. This week’s article examines the pitfalls of prioritizing outputs over outcomes and the lessons for modern teams. https://lnkd.in/ePq6cVbC

201

21 Comments

Creating a Detailed Project Plan and Timeline * Organized planning for efficient execution * Clear timelines for project milestones * Effective coordination among team members Learn more: https://lnkd.in/e3WK9X_N Read our playbook, "Software Development Playbook" written by Mark Turkel, PBSD Founder and CEO. This book was written for business owners and decision makers in a non-technical form, and will help you to thrive in the software development process. Available on Amazon. #PalmBeachSoftware #MarkTurkel #ProjectPlanning #Timeline

2

Being in product at a fintech is easy -> Users: "Let me explore before sharing my SSN"  You: "Makes sense!"  Compliance: "We need their SSN before they see anything” -> Growth team: "Can we remove the address verification step?"  Legal: "Actually, can we add three more verification steps?"  You: stares at conversion funnel -> Board: "Why isn't our onboarding as smooth as Stripe's?"  You: "Stripe has 200 engineers working on fraud"  Board: "But how hard can it be?" -> You: "We can verify users seamlessly"  Engineering: "6 month build minimum"  Sales: "We promised it to clients next week" -> Marketing: "Users love our simplified flow!"  Data: "40% drop-off at SSN screen"  You: "Have we tried..." -> Every stakeholder: "Just copy [insert big fintech]"  Reality: They spent 3 years and $50M on onboarding  You: opens LinkedIn jobs To every product leader balancing security and conversion: we see you 🎯

39

1 Comment

Why agents lack practical chops? They are missing - tool awareness - tools are not yet designed for agentic use (most are designed for human-in-the-loop) - hence RPA is such a big use-case for agents (programmatic tools / API based tools) - tool discovery - no universal registry - a lot of tools are human gated (think sales teams and demos) - tool provisoning - need access to buy and self-provision tools most agentic use-cases hence showcase demos like - flight booking - calendar booking / scheduling - social media posting - content research - tools that are ungated and high degree of programmatic or standardized interfaces. Looking at building universal agents - gotta solve the realtime vision + action challenge - there needs to be a model that can see pixels and manipulate browser / computer in realtime. I see this as the waymo or tesla moment for the digital world. You need the digital equivalent of lidar for computers. This will be the base model on top of which specialized agents can emerge - fine-tune the base model with training for tool use + reasoning systems will emerge. The agent stack for November is as below - building something in agents - HMU #agents #ai

32

1 Comment

I have to write my first technical blog in partnership with #Memoria Chief Architect, Victor Smirnov - explaining why DTR (DevSecOps Technology Research) is making a huge bet on HW/SW Co-Design Frameworks, which will blend beautifully with our #HLIX Hybrid Cloud Platform as a Product Strategy. There are numerous limitations with von-neummann architectures as described by professor Jeff Turner in our weekly #Abravera discussions to enable higher efficiency in symbolic AI - to enable #ComputationalConsciousness for distributed applications & robotics systems... Alexander Bothe. If AI Native Distributed Software Platform Strategies & Rethinking the 'brain' of a datacenter & the edge excites you, please reach out. The venture community needs the collective we've built now more than ever. Thank you Jordan M. Hutchinson for the invitation to https://lnkd.in/enpNaaWc for us to share some of this research DTR (DevSecOps Technology Research) is conducting in the space around cloud, and AI-Native Frameworks. The Future Is Open. There are no buzzwords here. We will execute. We will disrupt the status quo. If any investors want to join me in Salt Lake City UTAH before we close up this round, shoot me a DM. Also DM Me if you need current gen H100's for current gen inference & model training, served by our strategic partner, #NovoNimbus. #NovoNimbus || DTR (DevSecOps Technology Research) || #Memoria #Docker #Kubernetes

8

2 Comments

TRIPLE BAM! If you missed it, here's the recording of our live Q/A with Jay Alammar and Joshua Starmer PhD! We talked about many AI topics, answered lots of great questions, and talked about two great books: Hands-On Large Language Models: Language Understanding and Generation, and The Illustrated Guide to Neural Networks and AI!!! https://lnkd.in/dcPHs7YQ

44

1 Comment

When there is great uncertainty, there can't be predictability to any useful level. We have to reduce uncertainty to where it can be managed effectively, and we do this by running experiments... trying things. Does the stakeholder like what we've done? Does the customer like what we've done? Are we solving the problem that customers have? This puts us in the 'complex' quadrant of the Cynefin Model, where we have to try and (perhaps) fail, perhaps repeatedly, until we know enough. Yet, we still have a goal in mind and we at least know the first few things that need to get done... or the first thing we need to test to determine feasibility. Apple knew that unless it could source a ruggedized hard drive that would withstand being carried and occasionally dropped, the iPod would not be successful... so getting a commitment from Western Digital (who had been working on such a drive and was almost ready to ship) was the key decision point. The Wright Brothers knew that unless they could figure out a way to turn their airplane without 'skidding' and stalling a wing (a problem that had killed many of their inventive contemporaries) they couldn't build a safe airplane... oh, and find a lightweight, powerful-enough motor while you're at it. It all comes down to reaching decision points: do we know enough to know whether to continue, or whether to cancel our effort? In the example below (waterproofing a below-ground basement), we are going to have to continue. Therefore, we want to continue only at the pace at which we are certain of the decisions we're making. Is what we're doing working? If so, do more! If not, do something else. That is exactly what the contractor did in the post below, making 'MVP' experiments until the path to success was clear, until they had the 'Aha!' moment. So, when do we get predictability with useful accuracy? At the 'Aha!' moment. Figure out what 'Aha!' moments are needed, devise experiments to reach those moments and drive enough uncertainty out (you'll never get to zero) to where the range of uncertainty is acceptable. #predictability #forecasting #estimation #commitment #agile #planning #riskmanagement #projectmanagement

3

3 Comments

Very exciting era for curious minds, small businees owners and new entrepreneurs: 1. Tech / Engineering is no more a bottleneck for building quick prototypes. Use bolt new 2. For tougher problems and managing multiple files of code use cursor , use Claude 3.5 sonnet for coding 3. For online research, validating your ideas, competitor analysis - use perplexity 4. For working on your custom data, think PDFs / docs / notion - create simple RAG systems - use lang flow / flowise ( still early days, and glitchy software but they will evolve soon ) 5. For working on specialized tasks that can work on single shot prompt use AI Agents - try crew AI and pydantic AI. Very early stage of evolution - the biggest use case would be agentic memory but did not see a good example of this so far. #AItools #LLMs #RAG #AIAgents #Smallbusiness #entrprenuers

16

1 Comment

🇪🇺 Be part of the conversation on July 11th at 3:00 PM CEST! Our #NonprofitPulse2024 webinar features a rockstar panel discussing the latest nonprofit trends across fundraising, data security, and the development of AI. Save your spot today! #Fundraising #AIforGood #NonprofitTrends

2

Have you heard the news? Our team at Pendo.io is excited to be offering our Product-Led Certification Course for free! The course covers the fundamentals of being product-led and provides actionable product-led strategies to drive growth for your organization. Register now while it's still free to take the course and earn your certification! ⬇️ #ProductManagement #ProductInnovation #BusinessLeaders #Product

5

Scaling a business effectively often requires more than just adding new hires. The traditional approach of growing teams to drive growth can lead to increased communication challenges, slower decision-making, and diminished productivity. Instead of focusing solely on expanding headcount, it’s crucial to optimize existing resources and leverage technology. AI, for example, can automate repetitive tasks and enhance efficiency, allowing smaller teams to achieve more without the added complexity of a larger workforce. By adopting a leaner approach and integrating smart tech solutions, businesses can scale effectively and drive impressive results. #BusinessGrowth #AI #Efficiency

3