iiinova — where your digital self keeps evolving

nnenna hacks @nnennahacks.com · 19/05/2025

How to end your Sunday: coding 🙏🏽

nnenna hacks @nnennahacks.com · 28/03/2025

This AI workshop was incredible and glad I attended. So many inspiring women leaders looking to lead the charge in how to leverage LLMs effectively across communities. 🙌🏽

000

nnenna hacks @nnennahacks.com · 12/03/2025

vLLM East Coast meetup at Google 👍🏽 Met some folks from Red Hat’s AI teams and an architect from Cisco 👩🏽‍💻 Diverse, super technical crowd.

000

nnenna hacks @nnennahacks.com · 07/03/2025

I attended Harvard Data Science Initiative 2025 workshop on Responsible AI and Governance. 🤓 All I can say is WOW. 3 hours of pure intellectual insight & discussion. 🙌🏽 I got to ask THE Pat Gelsinger what he thought about synthetic data generation for LLMs. I’m still shaking from that moment. 😆

000

nnenna hacks @nnennahacks.com · 28/02/2025

What my tech meetup calendar looks like over the next couple weeks:

A meetup calendar with 4 events listed that I’m attending!
1. Google AI Seminar (Virtual)
2. International Women’s Day 2025 Boston
3. AI Meetup (March): GenAI, LLMs and Agents
4. Boston AI User Group March Meetup

010

nnenna hacks @nnennahacks.com · 20/02/2025

New book, new adventures 👩🏽‍💻🤖

A photo of a book titled “Generative AI on Google Cloud with LangChain” that arrived in the mail.

020

nnenna hacks @nnennahacks.com · 20/02/2025

Learned more about RAG, TAG, and BI at Microsoft last night hosted by the Boston AI User Group. I was particularly intrigued by the concept of “Semantic Cache” in LLMs. More research to do on this later 👩🏽‍💻

At Microsoft, a presenter with a slideshow is presenting on the “holy trinity of data analysis” where he has a diagram about all the systems working together in specialized LLMs : RAG, TAG, and BI.

230

nnenna hacks @nnennahacks.com · 19/02/2025

And here's the response from Claude 🔍👇🏽

A screenshot of Claude's response from my prompt.
"Claude Prompting to Understand Target Audience for Technical Content Creation":
Based on your LinkedIn content about AI vs. Automated Workflows, here's my analysis:
1. Ideal Target Audience
Your ideal target audience would be:
- Mid-level IT professionals and developers DevOps engineers, system
administrators, software developers)
- Technical managers and team leads making technology adoption decisions
- Business analysts and project managers interfacing between technical and business teams
- Technology consultants advising on digital transformation
- Non-technical professionals interested in understanding AI implementation vs. automation
2. Why This Is the Ideal Audience
This audience is ideal because:
- They're actively navigating decisions between implementing AI vs. traditional automation
- They need practical frameworks to explain technical differences to
stakeholders
- They're likely responsible for evaluating technology solutions with budget implications

Continued Response (part 2):

- They appreciate concise visual comparisons that clarify commonly confused technologies
- They're positioned to implement hybrid approaches in their organizations
3. Why This Content Works Well for LinkedIn
This content is excellent for LinkedIn because:
- It's visually engaging with a clean, branded design that stands out in feeds
- The slide-by-slide format suits LinkedIn's scrolling experience
- It presents complex technical information in digestible chunks
- It demonstrates your expertise while remaining accessible
- It addresses a timely topic AI vs. automation) that generates discussion
- The closing question invites engagement and comments
4. Areas for Improvement
- Add specific industry examples
Reason: Concrete examples make concepts more relatable
Value: Helps readers visualize applications in their specific sector
- Include cost/complexity considerations
Reason: Decision-makers need implementation context beyond technical
differences
Value: Addresses practical concerns that influence technology adoption
- Incorporate metrics or statistics
Reason: Data points add credibility and urgency
Value: Strengthens your position as a knowledgeable resource
- Add a "When to avoid" section for both technologies
Reason: Balanced perspective demonstrates deeper understanding

Continued Response from Claude (part 3 - final part)

Value: Helps prevent implementation mistakes and builds trust in your
expertise
- Feature a mini case study
Reason: Real-world implementation stories are compelling
Value: Demonstrates practical application of the concepts

The content is already strong, professional and well-structured. These
improvements would enhance its practical value while maintaining its visual appeal and digestibility.

010

nnenna hacks @nnennahacks.com · 19/02/2025

Here's my prompt. Notice the structure of the prompt: - uploaded content for context 📈 - stated a goal 🎯 - defined how to achieve that goal 🪜 - mentioned details on expected output 📝 - provided extra context for safe measure 💡

Screenshot of uploading a PDF document about "AI vs. Automated workflows" to Claude and providing this prompt:

"I need to know who my target audience could be for my technical content creation for LinkedIn.

Give me a summary of what the ideal target audience is for the PDF I uploaded.
Summarize why that would be the ideal target audience for this content.
Provide feedback on why this content would be great for LinkedIn.
Provide feedback on areas for improvement in this PDF. Explain the reason and value of the improvement suggestions, too.
---

For context, the PDF I created and uploaded is about "AI vs. Automated Workflows". I described their similarities, differences, best use cases for each, and use cases for a hybrid implementation."

110

nnenna hacks @nnennahacks.com · 17/02/2025

Not stepping outside this winter. Catch me indoors training the body, the mind, and some ML models 🫡

Mirror photo of myself in my living room, giving a thumbs up. Couch in the background with a yoga mat on the floor.

010

nnenna hacks @nnennahacks.com · 17/02/2025

Learning ML & AI engineering now as someone comfortable with Python feels encouraging. 🙏🏽 All those years of writing code in Python has lowered the barrier of entry to hands-on ML & AI engineering practice (for me), since it’s such a widely used programming language for these areas.

"A screenshot of a Jupyter notebook showing Python code for implementing Masked Self-Attention. The code defines a class 'MaskedSelfAttention' that inherits from 'nn.Module'. The notebook shows initialization code with linear layers for query (W_q), key (W_k), and value (W_v) transformations. On the right side, there's an annotation explaining that the masked_fill() method replaces True values with -1e9 in an attention mask tensor. The interface shows typical Jupyter notebook controls including a file menu, run button, and cell type selector set to 'Markdown'."

120

nnenna hacks @nnennahacks.com · 17/02/2025

Since I have the day off, I'm learning about attention mechanisms in transformers using #PyTorch today. 👩🏽‍💻 @deeplearningai.bsky.social

A screenshot from a DeepLearning.AI presentation comparing Encoder-Only and Decoder-Only Transformers. The image shows two diagrams side by side - one in pink (Encoder-Only) and one in blue (Decoder-Only). Both show the architectural components including Word Embedding at the bottom (with example text 'Pizza is awesome great <EOS>'), Positional Encoding, Self-Attention (regular for Encoder, Masked for Decoder), and Context Aware Embeddings/Generative Inputs at the top. Text in the middle explains that a small difference in using Self-Attention vs Masked Self-Attention has a profound effect on the capabilities of these transformer types.

140

nnenna hacks @nnennahacks.com · 17/02/2025

I’m worried about this. I use many of these phrases naturally in speaking AND writing and have been since I was a kid submitting essays in school. AI is trained on HUMAN-created words & writing styles, NOT the other way around. Don’t assume these formal phrases are AI. 👇🏽

2013151

nnenna hacks @nnennahacks.com · 13/02/2025

So far, so good? #CrewAI, #Gemini flash model, Google SERP AI, and other tools for building AI agents. A lot more refinement ahead to create sophisticated processes. 🤷🏽‍♀️👩🏽‍💻

100

nnenna hacks @nnennahacks.com · 11/02/2025

My calendar is filling up with tech community events. 🥹 Looking forward to these…

010

nnenna hacks @nnennahacks.com · 11/02/2025

Update: I installed #crewAI and generated a #Gemini API key. Let’s see if I can create DevRel AI agents 🧐

000

nnenna hacks @nnennahacks.com · 09/02/2025

I updated the design and pushed to main branch, but struggling with the Netlify deployment. My site might be too old and vanilla for this…?

000

nnenna hacks @nnennahacks.com · 08/02/2025

Code & Coffee in Cambridge today! 👩🏽‍💻☕️

000

nnenna hacks @nnennahacks.com · 04/02/2025

Last night, I installed Red Hat’s open-source AI project #InstructLab, downloaded some models, and ran the chat locally. 🙌🏽 From a methodological perspective, I’ll read the MIT-IBM research paper on LAB (Large-scale Alignment for Chatbots) to better understand the project’s underlying mechanisms.

140

nnenna hacks @nnennahacks.com · 31/01/2025

I read a chapter of “Infrastructure as Code, Patterns and Practices” by Rosemary Wang last night. Purchased on Manning.com Loved it. So clear and thorough. TBC.

Screenshot of Infrastructure as Code book by Rosemary Wang on manning.com. This is a live book where I can read and access it fully online.

110

nnenna hacks @nnennahacks.com · 28/01/2025

I did a 30 minute lower body workout and then 10 minutes of Pilates abs. Time for bed 🙏🏽

020

nnenna hacks @nnennahacks.com · 26/01/2025

Python, Tensorflow, Vertex AI, BigQuery, GCP things.

010

nnenna hacks @nnennahacks.com · 25/12/2024

Merry Christmas from Pippin 🎄❤️

010

nnenna hacks @nnennahacks.com · 20/12/2024

Returning to this book to revitalize my perspective on key success metrics in DevRel. 📝

Photo of a book titled “The Business Value of Developer Relations” by Mary Thengvall.

110

nnenna hacks @nnennahacks.com · 16/12/2024

‘tis the season 🎄🥂

040

nnenna hacks @nnennahacks.com · 11/12/2024

Built my new desk and I like it so far. Simple and sturdy. 😌

Image of my dark brown wooden work desk.

010

nnenna hacks @nnennahacks.com · 09/12/2024

I think he’s happy. 😸

Image of a black pet cat curled up on a bed sleeping.

020

nnenna hacks @nnennahacks.com · 04/12/2024

Stopped by my favorite consignment shop today. 😄

010

nnenna hacks @nnennahacks.com · 15/11/2024

photo ops from a networking event. these came out alright! 👍🏽☺️

010

nnenna hacks @nnennahacks.com · 11/11/2024

(2/2) There are costs beyond human effort to innovate tech. Check out the snippet from the report and compare it to the book’s humbling perspective on technology. Honestly, I'm glad that people are thinking about tech from a worldview and not in isolation.

There are costs beyond human effort to innovate tech. Check out the snippet from the report and compare it to the book’s humbling explanation of technology.

Honestly, I'm glad that people are thinking about tech from a worldview and not in isolation.

Text discussing how technology, while powerful, is ultimately limited by the laws of thermodynamics. It highlights the irony that as technology becomes more complex, we forget its dependence on finite energy resources and its tendency to increase disorder.

010

nnenna hacks @nnennahacks.com · 10/11/2024

Sunday hiking in Melrose, MA 📍

An image of beautiful trees in the forest during a hike with the sky in the background.

010

nnenna hacks @nnennahacks.com · 10/11/2024

Wonderful Saturday. Hot yoga at Equinox. Followed by lunch with a close friend. Got my watch resized. And got in a lot of steps browsing on Newbury Street. ☺️

A photo of Nnenna at the Rag & Bone store in Boston!

020

nnenna hacks @nnennahacks.com · 09/11/2024

For example, uploading an image like this into Gemini and asking it to provide ALT text I can copy/paste.

“The emperor isn't wearing any clothes!" This is the way one feels at the first recognition that technology really is naked and all of the artifice that surrounds it and what it is, is removed. All of it is a transformer. Every technology ever crafted, from the first stone tool to the most sophisticated microchip, is an expression of humankind's striving. In the process of transforming energy from nature's flow through culture and the other human artifacts, the energy is forever equilibrated. At the other end of the flow, it is used in a nonequilibrium state to sustain life (and the system where life is) for a fleeting moment. And the human has availed its future as technology has become more complex and its energy demands more profligate. It is ironic that as technology has become more disparate and unavailable for its domain, it has become more complex, something from scratch, through which we have come to see it as adding to the dependence of nature, as if it were generating its own energy independent of the world, some mysterious process, were more complex than the existing energy source, technology out of thin air, as if it exists first. The fact is, technology never creates energy, only uses up existing available energy. The larger and more complex the technology, the more energy it uses. As technology gets more complex and impressive, the more technology might sometimes appear to the operatives under the supreme reign of it, the uses and awesome, and impressive, as our available energy is used up, first, all matter-energy in the world is constant. It can be neither created nor destroyed, but only transformed from one state to another, just like everything else in nature. Those laws from an available to a dissipated form, or from an ordered to a disordered state. Second, the transformation of energy is always from an available to a dissipated form, or from an ordered to a disordered state. Technology is the transformer—nothing more, nothing less.

120

nnenna hacks @nnennahacks.com · 07/11/2024

He lives with me and thinks everything in the house is his 🥲

010

nnenna hacks @nnennahacks.com · 07/11/2024

Pippin loves hydrangeas! 💐🐱

010