Ofer Mendelevitch

"Productization is an affirmation," writes Adrian Bridgwater, "when any good or service evolves to a point where it can be productized and boxed, it denotes a certain solidity and branded consumability." In our conversation, I was delighted to share how #SAS equips organizations with readily deployable AI technology to productionize real-world use cases. https://lnkd.in/enUGCgAi #AIML #sasusers

9

My interview on The Post's AI collaboration with VTech in this article:

103

3 Comments

Kudos to Kshitiz Parashar for his pioneering work on Agentic Chunking! His latest primer has quickly become the must-read guide for AI Application Engineers looking to optimize LLMs for smarter chunking, better retrieval, and sharper accuracy. This Speaks volumes about the calibre of the team which I am super proud of.  Even a lean, dedicated group can now harness the power of LLMs to design state-of-the-art architectures—proof that AI is leveling the playing field like never before. Ready to dive in? Explore the full article here: https://lnkd.in/eJ8cem3x #aiagents #chunking

37

1 Comment

Regulatory oversight can provide short-term stability and protection, but if not carefully planned, it can often stifle innovation. The SEC's regulation of #ICOs in 2018 is a prime example. The increased regulatory burden and uncertainty led to many ICOs excluding US investors and positioning themselves as foreign offerings to avoid SEC jurisdiction. This shift limited access to capital for US-based early-stage startups and created new challenges for companies seeking to raise funds through ICOs. Most importantly, the lack of clear guidance made startups hesitant to explore new financing models, fearing regulatory repercussions. It's ironic that a #gambler can lose their house in minutes in #Vegas without the same level of scrutiny and regulation on #casinos! And kudos to Office of California Governor Gavin Newsom for not getting too excited or bowing to pressure! If the government wants to be part of innovation, they should offer grants and non-dilutive funds to startups, just like #Europe and #Canada do with - SR&ED Tax Incentive Program - AgriScience - NRC-IRAP - EU Grants for Digital Health and Technology Startups - IIA Investment Programs and many more! This way, they can keep an eye on startups and disruptive solutions as they submit their progress each year to qualify for more funding. Let's support our startups the right way! #InnovationBeforeRegulation #AI PBS KQED NPR BBC CNN

9

3 Comments

The future of AI may be as close as R2-D2 and C-3PO with the rise of small language models. #SLMFuture #embodiedintelligence #pocketai #aiagents #cambrianexplosion #humanaiconnection

2

Interesting discussion and episodes on the use of the AutoGen framework for Agentic Workflows at the Microsoft Research Forum. The #Agenticframework allows LLMs to allow to perform more complex and intelligent tasks autonomously through one of the key concepts of reflection and autonomy. It has significant impact on practical applications such as personalized medicine, automated travel concierge, autonomous navigation etc.. It was also very interesting to read blogs posted on deeplearning.ai by Andrew Ng. https://lnkd.in/daa28seD

10

2 Comments

Here's a needed reminder (and warning) from FinTech Futures about how commonplace #deepfake scams are becoming as large language models (#LLMs) and #AI tech improves. AI-generation is becoming sophisticated at replicating human-like speech. Phone scams, fraudulent videos, and other types of scams are going to become more convincing to those who are unprepared. This article also highlights FinVolution, a fintech platform that has a focus on anti-fraud technologies in the Asian region. Right now, algorithm scientists are trying to spot #hoaxes by spotting the nuances and imperfections that come with AI voice cloning technology, basically digital forensics to help customers identify the artificial. I do truly mean it when I stress that AI, like other forms of technology, is ideologically neutral. In the same way that plenty of smart innovators are utilizing AI for good, there are also bad faith actors misusing AI for their own nefarious purposes. So, even though we live in an exciting time for generative AI, lets also remember to practice caution and use our better judgement to protect ourselves from #scams, #misinformation, and #fraud caused by deepfake technology.

12

5 Comments

In a recent conversation with Sean Falconer on the Partially Redacted podcast, we discussed the widespread challenges of leveraging regulated data and how the Integral team is tackling this challenge while enhancing underlying data. Healthcare data is often plagued with inaccuracies, inconsistencies, and compliance risks. This includes things like: - Misspelled medication names - Incorrect patient genders - Data in the wrong columns - Sparse or missing values - Protected health information (PHI) We've developed an end-to-end suite to tackle these challenges head-on: 1. Automated compliance to rapidly certify and transform datasets  2. Continuous monitoring to compliantly deliver transformed, real-time data from source to destination 3. Intelligent dataset enrichment to detect quality issues and recommend high-value data sources to fill gaps 4. Seamless partner integrations with data providers, originators, and analytics platforms Thank you Sean Falconer for a great discussion. Listen to the full episode here: https://lnkd.in/gBKjPkKM #healthcaredata #compliance #dataenrichment

24

Over 120 people tuned in to watch our live webinar together with Kern AI's Jason Wirth and Leonard Püttmann on Tuesday talking about Entity RAG and LLMs. We showed how the combination of GPT 4, plus customer data hosted in Tilores, all build together using Kern AI's cognition platform, can give a superior outcome versus just using a vector database alone. If you missed the webinar, don't worry - you can watch the recording below on YouTube where we use an insurance industry case study. https://lnkd.in/d8BHpRm6 #LLMs #ChatGPT #RAG #datascience

14

1 Comment

Interesting read, thanks for sharing James Leckie https://lnkd.in/gtktGhjM Excellent deep dive analysis on the current state of AI tackling the nuances of pedagogy and the limitations in present base models. "Human tutors excel in understanding real-world context and personalizing lessons based on individual learners. LearnLM, as a non-embodied AI, lacks deep contextual awareness and the ability to fully personalize or interpret non-verbal communication." Such a fast moving space!

7

The Three Components of the Unstructured Data Stack by @AstasiaMyers at @felicis (great to see @falkordb listed as critical foundational data infrastructure) https://lnkd.in/dUs3tYCR

7

NYC Data Science Academy today from 1:00-2:00pm EST will be our weekly Live Learning: The Surprising Effectiveness of Test-Time Training for Abstract Reasoning. This is a https://lnkd.in/dWfxd8fW from MIT that came out on Monday which points to some very interesting and promising advantages of the "test-time compute" paradigm that we have seen with models such as OpenAI o1. While this is less of an off-the-shelf technique that anyone can apply right now, it is something that, with a bit of effort, you could play with when using open source models yourself. Our focus will however be in understanding what this is, and why it is impactful. Some of the topics we will go into are: - self-supervised learning - https://lnkd.in/grymPAKi - https://lab42.global/arc/ and why it a) is an important benchmark and b) why it has been https://lnkd.in/dphV2PUB so far Join Zoom Meeting https://lnkd.in/eScKr6kJ Meeting ID: 946 1359 3549 Passcode: 100607

9

Part II of our blog is out! In this installment Abde Tambawala and I explored usage metrics and pricing models across different AI modalities and discussed how traditional User-based subscriptions will need to be supplemented with usage metrics. Take a look and let us know what you think! #pricingstrategy #ai #monetization

27

2 Comments

Generative AI is stepping into the role of a true virtual assistant in healthcare 💻 Anthropic’s new API takes AI to the next level, enabling it to interact directly with computer operating systems and applications. This breakthrough allows AI to manage files, navigate software, send emails, and even extract information from documents—tasks that previously required human intervention. For healthcare professionals, this could revolutionize daily operations: ▶️ Streamlined workflows: AI can open medical records, organize patient data, and generate reports. ▶️ More time for patients: By automating administrative burdens, healthcare teams can focus on delivering care. ▶️ Reduced burnout: Letting AI handle repetitive tasks could ease the workload on healthcare staff. However, as AI gains this level of access, security becomes paramount. Managing sensitive patient data means operating within a HIPAA-compliant framework to protect privacy and maintain trust. This evolution highlights the potential for generative AI to shift from a passive tool to an active participant in healthcare operations, promising more efficient workflows and improved patient outcomes. What other use cases do you envision for computer use in streamlining healthcare? And how can we ensure that innovation remains secure and patient-centered? #HealthTech #AIinHealthcare #GenerativeAI #HealthcareInnovation

28

2 Comments

How AI changes data landscape? Story behind DataChain:

13

"This will be one of the great human projects of our time." Alexandr Wang is spot on. Frontier data is the next gap to address as GenAI adoption matures. Where I disagree is the source of that data. The best data won't come from AI labs and labeling platforms like Scale AI. When solving for QUALITY (not quantity) of data, proximity to UX is important. Data brokers and offshore labelers just can't replicate that. What I've seen in the field: The strongest AI teams closely oversee data labeling (in-house or managed contractors) to make sure it captures the nuances of their use case. The application layer will play a crucial role in this great human project.

13

2 Comments

Almost twenty years ago, in 2000, Eric Brewer introduced the concept of a fundamental trade-off among consistency, availability, and partition tolerance. This concept, now known as the CAP Theorem, has been instrumental in evaluating distributed system architectures for decades. With the rapid proliferation of streaming platforms, there's a growing need for a similar model to evaluate the trade-offs between Cost, Availability, and Performance in data streaming platforms. In response, we introduce the New CAP Theorem tailored for cloud-based data streaming systems. The New CAP Theorem is designed not to rank systems but to emphasize the importance of making informed choices based on a balanced consideration of Cost, Availability, and Performance. We encourage everyone to utilize the New CAP Theorem to select the most suitable technology from various data streaming platform options, including #ApachePulsar, #ApacheKafka, StreamNative, Confluent, and many others, to meet your unique business requirements on your data streaming journey. https://lnkd.in/gcuqViBm

46

4 Comments

With the release of o1, a model with advanced meta-thinking, using a "mixture of agents" becomes even more appropriate. For instance, o1 can now be used to plan prompt sequences to be performed by less expensive or task-specific models and tools. Meta-prompt engineering is now officially a thing. https://lnkd.in/eTqwp_mt

17

1 Comment