Stephen Williams

Interesting Autonomous car behavior. You'd think that GNSS would at least inform the car of what lane they are in and direction of traffic, while the camera/lidar would determine that all the other cars are pointed the other direction...

6

1 Comment

Do you want to be more productive? We can design best in class AI agents for any business problem. This article is a small piece of larger echo system which demonstrate the capability of Agentic systems. stay tuned to my blog space for more articles like this!! Qdrant, crewAI, LangChain, Ollama

16

May brings us a bevy of recognitions that, given the current state of world affairs can seem heavily at odds. Many know May as Asian American, Pacific Islander, and Hawaiian Heritage Month, but it's also Jewish American Heritage Month and Military Appreciation Month. As we watch events unfold in the Holy Land and see passionate activitism at home and abroad meet stern and violent opposition and generate understandbly heated discussions about American diplomatic and military strategies in Isreal and elsewhere, it can feel like expressing mutual support and appreciation for our military, Jewish people, and Palestinian self determination are impossibly conflicted. However, if are to succeed as a cross-cultural, multi-racial, and principled society based on freedom and equality, that should not and cannot be the case. We must resist reactive and siloed perspectives on these issues and acknowledge the nuance of the situation. We can honor and appreciate the sacrifice of our individual soldiers and support personnel and their families even if we don't agree with the way our national leadership deploys them and funds foreign allies. We can honor and amplify the culture, voices, and history of the Jewish people in America and around the world, and we can combat and stand against Antisemitic statements and sentiments while still divesting from the actions of Israeli leaders. We can speak and protest in support of Palestinian rights, safety, and legitimacy, and call for ceasefire without condoning the tactics of Hamas or other groups. People too often dismiss DEIA efforts and identity recognitions as "divisive" but they aren't inherently, and if we are to succeed sustainably as a nation and a society, they CANNOT be. We MUST speak boldly to the experiences of specific marginalized groups and work collaboratively toward solutions that seek peace, stability, and security for all people. Such a strategy benefits everyone. If we refuse to acknowledge identities and we refuse to hear the concerns of specific groups, then we only prolong the motives that underpin social strife and breed conflict. A lasting solution demands comprehensive, Affirmative, deliberate, and cooperative action. ✊🏿✊🏾✊🏽✊🏼✊🏻 #DiversityMatters #AAPIHeritageMonth #JewishHeritageMonth #MilitaryAppreciationMonth

1

We (as a planetary civilization) haven't even mastered basic CTP-based technologies or developed to a level of utilizing CTPtech in day-to-day living. 'What' precisely do we have to offer advanced civilizations? Here's a list... Social justice (misuse of) Cultural Marxism Marxism Crony-economics Tribal/identity politics Righteous indignation/political outrage Mass homeless Mass starvation Crime Violence Social disparity Depression Suicide Murder Political corruption War Spiritual ignorance or understanding the deeper true nature of reality. (insert any negative social issue...) We have little to offer them until we are much further along. If it weren't for the fact what CTPSci Sciences can offer humankind, I wouldn't be posting about going on 3 years now. If Drazen were still here and the world had NOT ignored the next #heiroftesla, I wouldn't have be here telling you all about it! I'd likely be doing something else. THE OTHER FACT OF THE MATTER - I WILL REPEAT AGAIN AND AGAIN THEY have not contacted us - BECAUSE WE HAVE NOT FIGURED OUT 'how' to contact them! Advanced civilizations are NOT using radio waves or P-energy to communicate over universal medium. They use CTPSci and #CTPtech to do this! Quite frankly, this article 👇 sucks as well as its entire premises and suppositions. (https://lnkd.in/egFcu-Cz) #newscience #newworlds #newcivilization #wakeup

5

Just obtained the OMG Certified Modeling Professional - Model Builder Fundamental cert! Fun preparing for this one, I really gained some new perspectives on the SysML language.

9

AI has always been more than a buzzword to me, it is my life’s passion. Since grad school it has been at the forefront of my mind, where I started with humble neural networks trying to discern the underlying patterns. Now, AI is not just an academic interest; it's the heartbeat of my strategy. In our latest blog post, I unveil the AI roadmap that is propelling us into the future. At the heart of Sierra Space's AI strategy lies a trifecta of AI-oriented domains. The first domain is commercial solutions—spurring efficacy and speed today. These are tools ready for immediate deployment, from general AI assistants to honed experts, each designed to make our operations faster and more efficient than ever before. The second pillar stands upon our profound domain knowledge paired with unique data. Here, AI transcends its conventional role and ventures into the realm of insight generation—sifting through vast oceans of data to uncover the pearls of actionable intelligence. By leveraging our historic datasets, our algorithms are equipped to fuel strategic forecasting and prompt incisive decision-making, thereby uncovering pathways previously untraveled. The final pillar is our third and most exhilarating domain: mission-specific applications. Think Dream Chaser spaceplane, where AI is our guide through orbit, safeguarding our global interest and providing an autonomous edge in real-time environmental monitoring and tactical decisions. It is this facet of AI that crafts the sharpest edge of our competitive sword—ensuring our partners and customers achieve the previously unachievable, confronting unique challenges head-on with unparalleled AI capabilities. Dive into our latest blog post to discover Sierra Space's vision for an AI-driven future. https://lnkd.in/gW63ZZAX

79

1 Comment

How is Humes "is-ought" problem associated to AI? Hume's guillotine states that one cannot get to a moral conclusion based on just facts alone. For the AI variation - Given a situation no one decision can always be considered intelligent -> This is both true for a human and a machine. So when designing a system with AI make sure you have Hume's guillotine in mind to constrain the reasoning and paths that can lead to failures caused by divergent AI reasoning. https://w.wiki/9wkD

2

Cool new ‘@’ feature in Gemini (LLM alternative from Google). Asked about the top 5 videos on YouTube to learn about Prompt Engineering and a summary of the first one. Gemini’s 😎response: “Sure, here is a summary of the first video 4 Methods of Prompt Engineering by IBM Technology: The IBM video explains four different approaches to prompt engineering to get better results from large language models (LLMs).: Retrieval-Augmented Generation (RAG), Chain of Thought (COT), REACT, and Directional Stimulus Prompting (DSP). * 𝐑𝐀𝐆 (𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥-𝐀𝐮𝐠𝐦𝐞𝐧𝐭𝐞𝐝 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧): This approach involves giving the LLM access to your domain-specific knowledge base. This helps the LLM to avoid hallucinations, which are incorrect results that can be generated when the LLM is trained on a massive dataset of general information from the internet. For example, if you ask an LLM directly about the total earnings of a company in a specific year, it might give you a random number because that information is not likely to be reliable in its general training data. But, if you use RAG and give the LLM access to your company's financial data, it can provide a more accurate answer. * 𝐂𝐎𝐓 (𝐂𝐡𝐚𝐢𝐧 𝐨𝐟 𝐓𝐡𝐨𝐮𝐠𝐡𝐭): This approach involves breaking down a complex question into a series of simpler steps. This helps the LLM to understand the reasoning behind the question and provide a more explainable answer. For example, instead of asking an LLM directly for the total earnings of a company in 2022, you could break the question down into steps like: 1. What are the total earnings of the company for software in 2022? 2. What are the total earnings of the company for hardware in 2022? 3. What are the total earnings of the company for consulting in 2022? By following these steps, the LLM can arrive at a more accurate answer and explain how it got there. * 𝐑𝐄𝐀𝐂𝐓 (𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥-𝐀𝐮𝐠𝐦𝐞𝐧𝐭𝐞𝐝 𝐂𝐨𝐧𝐝𝐢𝐭𝐢𝐨𝐧𝐚𝐥 𝐄��𝐜𝐨𝐝𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐞𝐫𝐬): This approach is similar to RAG, but it also allows the LLM to access public knowledge bases. This can be useful if you are asking a question that requires information that is not in your private knowledge base. For example, if you want to compare the total earnings of a company in 2010 and 2022, you might not have the 2010 data in your private knowledge base. But, REACT can allow the LLM to access public financial data to find the 2010 data. * 𝐃𝐒𝐏 (𝐃𝐢𝐫𝐞𝐜𝐭𝐢𝐨𝐧𝐚𝐥 𝐒𝐭𝐢𝐦𝐮𝐥𝐮𝐬 𝐏𝐫𝐨𝐦𝐩𝐭𝐢𝐧𝐠) This approach involves giving the LLM a hint about the specific information you are looking for in the answer. This can help the LLM to focus its response and provide more relevant results. For example, if you ask an LLM about the annual earnings of a company, but you are really interested in the earnings for software and consulting, you can use DSP to nudge the LLM on specific areas. #AI4Good #JTBD https://lnkd.in/ep8pRG4j

5

Learn how cloud-native PLM can help aerospace and defense manufacturers comply with the new DoD Digital Engineering Instruction 5000.97 and achieve a digital transformation.

2

Just finished the learning path “Building AI Literacy”! #artificialintelligenceforbusiness #artificialintelligence

1

US researchers propose adding a new traffic light to help autonomous vehicles (AVs) move in convoys, reducing congestion. Dr. Ali Hajbabaie of North Carolina State University explains that this extra signal would guide AVs through intersections more efficiently. Currently, AVs follow manual traffic, but the new signal would allow AVs to move in groups, cutting delays. Research shows a significant reduction in congestion with 30-40% AVs on the road. Manual drivers can still navigate by joining AV convoys under specific rules. MORE DETAILS: https://lnkd.in/dsd33SSq #AutonomousVehicles #TrafficManagement #SmartTrafficLights

1

K-GPT: A custom GPT based on [Zero To Hero] utilizing tiktoken with the intent to augment AI Transformer-model education and reverse engineer GPT models from scratch. #AI #ML #GPT #ReverseEngineering https://lnkd.in/egwpjaJk

3

Chapter 35: Functions, w/o Param, w/o Return 📷 Reverse at the Assembler level, learn to understand how functions w/o params and w/o return work in memory as we reverse engineer an embedded binary! #CyberSecurity #Embedded #Development #ReverseEngineering https://lnkd.in/enM_-gnM

4

Today I will be talking to the U.S. Naval Research Laboratory High Assurance and Software Testing and Evaluation team lead by Ramesh Bharadwaj. The topic will be the logic of unsupervised monitoring of noisy agents. That logic has to contend with the infinite recursion that judging the judges creates. The ubiquity of this problem is understood by the daughter of Monk's assistant in this episode from the namesake TV series. Who checks the level checking level?!! https://lnkd.in/ezRWfdr6

2

Advancing Defense and Manufacturing with Swarms of Large Language Models (LLMs) In an era where precision and efficiency are paramount, Swarm Corporation is pioneering the integration of Large Language Models (LLMs) into the defense and manufacturing industries. Our commitment to technological innovation is transforming core operations, enhancing strategic capabilities, and driving forward the future of manufacturing and defense. Explore the Impact of LLMs in Defense and Manufacturing: 👉 Streamlined Supply Chain Management: Employ LLMs to optimize logistics and supply chain decisions, reducing costs and improving time-to-market. 👉 Enhanced Equipment Maintenance: Implement predictive maintenance models that anticipate equipment failures and schedule repairs, minimizing downtime and operational costs. 👉 Robust Security Protocols: Leverage LLMs to strengthen cybersecurity measures, ensuring that sensitive data and operations are protected against emerging threats. 👉 Improved Compliance and Quality Control: Utilize AI-driven systems to monitor compliance with industry standards and government regulations, ensuring product quality and safety. 👉 Innovative Product Development: Accelerate the design and testing phases of product development with AI that simulates real-world conditions and outcomes, enhancing innovation and efficiency. Swarm Corporation is not just adapting to change; we are creating it. Our expertise in deploying LLMs within the defense and manufacturing sectors provides a critical competitive edge, ensuring that our clients are equipped with the best tools for success. We invite industry leaders to explore how our LLM solutions can revolutionize your operations. To learn more about our capabilities and how we can assist in achieving your strategic goals, please book a discovery call with us. 🔗 Ready to transform your operations? Schedule your 30-minute discovery call here: Join us in leading the charge towards a smarter, more efficient future in defense and manufacturing.

4

1 Comment