Passionate about applying leading-edge technologies for intelligent data discovery and collection, analytics, visualization, and reporting in business and industrial environments. Technical expertise encompass full SDLC skills from business and systems analysis, design, programming, debugging, deployment, support, and training. Favorite languages are Python, SQL, and web languages (HTML, CSS, Javascript) on Windows, Linux, Docker, and Azure/AWS/GCP Cloud environments (as required).

I actively research and develop applications, and training to harness Generative AI, Prompt Engineering, and intelligent workflow chaining applications using Large Language Models. I have experience with 100s of Python packages to develop applications and utilities for AI, intelligent advanced predictive analytics, ML, deep learning, and web-based systems. I have 30+ years experience an trainer and educator developing and delivering training curricula, courses, workshops, and short courses. I teach from time to time, but my passion is training people to use tools and methods to learn faster and remember for longer.

I love applying my business analysis and technical skills in large and small enterprises. I am an acknowledged SME expert in data and records management, IT systems, and scientific and lab informatics across value chains in pharmaceutical and biotech R&D through Manufacturing, Finance and trading. Contact me to engage my skills on your project.

In my personal time, I am actively learning and applying the State Of The Art (SOTA) Generative AI and deep learning tools and packages. I am working to SMARTLY SIMPLIFY advanced AI and data analytics. There are (too) many steep learning curves to climb before the masses can apply deeply AI technologies. TOO MUCH AI NOISE -- and NOT ENOUGH SIGNAL! People must discover first-hand what really works, and raise the baseline of AI wisdom high above popular media noise. My job as a trainer is to transfer hard implementation skills, and instill know-how in my trainees and clients. AI technology is advancing FAST toward "AGI". First it was the pitter patter of baby steps, but now the gallop has started, with leapfrogs every other week. Major breakthroughs in 2024 within months will cause seismic shifts in the balance of power between humans and AI agents.

Every week I am testing yet another LLM Agent builder application like OpenDevin, AutoGPT, MetaGPT, LocalGPT, DoctorGPT, babyAGI, DB-GPT, Pinokio, GPT-Engineer, AutoGen Studio, and other domain-specific tools. My goal is to learn Large Language Models (LLMs) as much as possbile and apply them to text, images, sound, and music. The goal is comprehensive and multi-modal generative experiences to accelerate human learning and individual power. One current project is to create a State-of-the-Art (SOTA) AI Guide and Developer Assistant for Prompt Engineering and Intelligent Agent Chaining Systems. New bleeding-edge tools appear weekly, far too fast for regular humans to consume or master in a normal workmonth. We need automated real-time knowledge acquisition tools that simplify explain new tools to newbies, "on demand" and "just-in-time" in the moment, before the spark of human curiosity extinguishes.

I have a strong commitment for individuals to run high-performing "democratized" LOCAL AI systems -- that is 100% CLOUD LOCAL -- without requiring ANY online cloud platforms. LOCAL AI systems must not be considered "edge" nodes on the global internet. What people call "The Cloud" is a few Big Tech cloud platform vendors WITH COMMON GOALS TO INSOURCE AND CONTROL AS MANY TECH JOBS, AND AS MUCH AI POWER AS POSSIBLE. The AI Arms race is to beat or consume competitors anywhere they arise. The risk is that Big Tech cloud companies are steadily and substantially in-sourcing jobs to their own corporations, leaving millions of citizens without meaningful work.

Citizens must re-power, re-source, and revert "The Cloud" back to "The Edge", or entire segments of white, brown, and blue collar workers will be marginalized by 2030. Once cloud companies "OWN" jobs, they can and will be placed jobs anywhere that quality labor is cheapest and highest quality. AI agents are an obvious endpoint to "in-source" jobs once cloud companies own them. AI agents require only electricity to work. AI agents don't need salaries, nutrition, nurturing, sleep, exercise, sex, emotional support, vacations, or other "human overhead". The argument that AI cannot produce the same high-level of quality as very intelligent and compassionate humans is simply wrong. AI already (often) produces higher quality and consistency for many tasks and work products. AI systems quality and sophistication are improving must faster than most people can comprehend. From the 2000s the re-engineering waves instigated by Mike Hammer and implemented by SAP moved millions of US jobs off-shore. In the mid-2020s, more jobs will move off-ground to the cloud, and in the early 2030s, many jobs will disappear to AI agents and robots. Humans left on "The Edge" will be marginalized. The Cloud moves economic power and control away from individuals and gives it to Big Tech. White collar IT managers, CIOs, CTOs, CDOs beware! ... AI is coming for your job much sooner than you think. Local, democractized AI must bring power back to the people.

Individuals must ensure that AI stays democratized and runs equally well locally -- local-first -- and move online when absolutely required to train AI models, then back to local devices. Distributed Tech (local clouds) must prevail in the best interests of individuals.

***Remember "Human Lives Matter" (HLM)! Get out of the clouds and put jobs back on Earth. Cloud computing is a tool, not the endpoint for society. ***

https://www.linkedin.com/in/rich-lysakowski-phd

Some of my favorite phrases are:

• "Talk is cheap, show me the code" ~ Linus Torvalds
• "Ship to Learn" ~ a GitHub Core Value
• "Always leave a place better than you found it." - Unknown
• "Everyone you meet is fighting a battle you know nothing about. Be kind. Always." ~ Robin Williams

{"Author": "Rich Lysakowski", "Updated": "2024.06.22" }