Activity
Mon
Wed
Fri
Sun
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
What is this?
Less
More

Owned by Sam

Systems of Impact

25 members • Free

Learn how to save a day per week using AI as a non-technical Product Manager.

AS
AI Systems

1 member • Free

Memberships

Tentmakers

3.2k members • Free

Dopamine Digital

4.5k members • Free

Skoolers

174.9k members • Free

The Receive Academy

1.3k members • Free

AN IDOL CALLED FOOD

255 members • Free

Tech Snack University

11k members • Free

Synthesizer

28.7k members • Free

Zero To Founder by Tom Bilyeu

1.7k members • $119/m

Scale Your Coaching

19.7k members • Free

8 contributions to Systems of Impact
Custom GPT Poll
Hey all, it's been a crazy week. My dad passed away this week, so I've been back home helping my Mom with everything. Grief will come in waves, and I'll experience things as they come. In the meantime, I will continue moving forward. With that, I'd love to know what folks want AI's help with streamlining. The winning option I will create a custom GPT for.
Poll
6 members have voted
1 like • Oct '24
Hey all, it looks like Market Research won. I'll work on a custom GPT to get started and make this process a bit easier.
1 like • Oct '24
@Jackie Seem — it depends on the version you use. We tried a Google Workspace version and I was highly unimpressed. It was pretty bad. GA version, it’s ok. Chat GPT has access to the internet and I find does a better job incorporating that information. A few weeks ago, Gemini kept referencing Reddit a lot still, and you can guess how that went. Overall, I use ChatGPT for most things, and the new versions of Claude 3.5 Opus for coding and development. I personally haven’t found any value to use Gemini other than the extremely large context window. I’m going to test out granola.ai with NotebookLM and see how that works for meetings.
Start Here!
Welcome! Thanks for joining Systems of Impact. I created this community for Product Managers that want to be more productive using AI. As we continue to grow, I will continue to add content, share ideas, and bring the latest and greatest in AI to help Product Managers like you. A few steps: 1. Skool is gamified. Post and Comment and have them engaged with will increase your level. As I release more programs and ideas, some may be behind these levels. Get started now so when they are released, you can dive right in. 2. Like this post. 3. Introduce yourself to the community. If you want to go fast, go alone. If you want to go far, go together. Get to know like-minded folks and build connections. In your intro, let's start with where you are in the world, why you joined the community, and something you like to do for fun. 4. Go comment on someone else's introduction. 5. Go to the Start Here course for a walkthrough of the platform. Again, welcome and I look forward to your success!
1 like • Sep '24
@Niels van der Linden — thanks for joining! My daughter is 3.5 and brings a lot of joy, among other emotions, lol.
Llama 3.2 Running Locally
Hey Folks, I just did a quick video walking through how to get Llama 3.2 running locally on your computer. I'll provide some links below to following along with, but I was able to get it up and running in about 5 minutes. I've also done a quick walkthrough how to get it setup. If anyone has any questions, drop them below. Ollama: https://ollama.com/ Ollama Llama 3.2 Page: https://ollama.com/library/llama3.2 Docker: https://www.docker.com/ Open WebUI: https://docs.openwebui.com/getting-started/
4
0
Llama 3.2 Running Locally
AI as a Thought Partner
We use CredalAI at my company and today we learned how we can create our own co-pilots within it. The demo was for writing customer-facing or cross-functional emails/documents but the demonstrator mentioned you could also use the co-pilot as a thought partner. I'm curious if anyone if anyone in the community has done this and what was your level of success?
2 likes • Aug '24
@Jackie Seem — Not really. The prompt framework is more geared towards a thought partner. LLMs are pretty good at summarization at this point. To your point, what you focus on may be different. In those cases, I’d give the LLM a role (I.e., product manager with 20 years of experience with a focus on go to market strategy) and then ask it to summarize. Once summarized, validate the output. The role you give an LLM will determine what it focuses on with the summary.
1 like • Aug '24
@Jackie Seem — TL;DR: Hallucinations cannot be avoided, merely mitigated. Context: I tend to use perplexity to measure accuracy in an LLM. To manage hallucinations, it will depend on your data, and the model. I’ll provide two examples. 1. Let’s assume you have a structured data set in a csv. You could fine tune 3 different roBERTa models with different hyperparameters, and feed those into another roBERTa model. Multimodal in this case helps reduce perplexity and leads to less hallucinations (they won’t be zero, but far less common). 2. You have a ton of unstructured data in the form of PDFs. You could use Llama 3.1 (LLM) and fine tune it. Or, if that data set is changing often enough, you can implement RAG. As it stands, RAG is pretty tough to implement in a production environment. Something to note, this is a GenAI issue and not really an ML issue. Here’s a link going over different measurement methods. https://research.aimultiple.com/large-language-model-evaluation/
Curious How to Use AI to Get to Senior PM Levels?
Getting promoted can be difficult, especially in today's market. That's why I created the DEER framework, and an assistant GPT. The framework goes over how to get promoted and the exact process I've used to increase my income over 240% in 5 years. In addition to the framework, I created an assistant GPT that can do the following: - Skill Assessment - Promotion Target - Gap Analysis It not only helps with the skills side of things, but I've included in the assistant GPT roughly 8 different company rubrics. It takes these into account when you do a skills assessment and a gap analysis. It can walk you through the process described in the framework. To be clear, it's not a quick win type of process. It takes deliberate thought and action. This program is unlocked at Level 2. See you in there!
5
0
1-8 of 8
Sam Montoya
3
34points to level up
@sam-montoya-4289
Helping non-technical Product Managers be more productive and deliver value faster using AI.

Active 11d ago
Joined Aug 15, 2024
Powered by