R to @kgmodi: Ohk , will try this out

no filter, no edits Colors in New Jersey during this time of the year are always mesmerizing.

RT by @MannyBuildsTech: They: "So your app is just an OpenAI wrapper." Me: "No, my app is an OpenAI token reseller." They: "What?" At its core, OpenAI wrappers act as token resellers. OpenAI wrappers interact with OpenAI's GPT-3.5 / 4 models by sending sequences of words to the model. Each token can be thought of as a chunk of data, often corresponding to a word or a few characters. These wrappers purchase tokens from OpenAI and then sell them to end-users, thus earning the title "token resellers." Just as refineries process crude oil to produce valuable products like gasoline and diesel, OpenAI wrappers take the raw tokens and refine them, enhancing their utility and value for end-users. Due to this enhanced utility, these "refined" tokens hold a higher market value compared to the raw tokens. The profit comes from the difference between the purchase price of the raw token and the selling price of the refined token. Keep building that OpenAI wrapper!

R to @kgmodi: Yes, there will always be heavy rains or drought-like conditions that can destroy a farmer in an instance. Adapting and getting back up goes without saying. Maybe you'll hunt for the time being, but you will never give up on farming

R to @sarangbhirad: Good analogy. But you also need to prepare for the day when Algorithm Gods decide to change the tide? For example you need to prepare when Google’s update causes significant negative changes to the organic growth.

R to @kgmodi: Love those uses Kunal. We had a long discussion on this last night. Overall I think all of this has the effect of making development far more enjoyable.

R to @Dj__: No plug-in just ChatGPT-4’s Advanced Data Analysis where you can upload data and start asking questions.

R to @kgmodi: Any plug-in was used ?

R to @kgmodi: Looking for next innovation & observation

Next level unlocked.

R to @LBacaj: Great example of using AI for not starting from scratch. I use a combination of ChatGPT-4 and Github Copilot and the friction to solve problems has decreased. ChatGPT is great at summarizing higher level concepts, error debugging and pointing you in a direction when you are lost. Copilot is fantastic when it come to autocomplete based on previous context. AI can get things so wrong - and so can developers. But if you use these tools as assistants and not as your replacement they work great - for willpower, debugging and getting things done.

R to @anillraju: yes, the size limit is 250mb.

R to @kgmodi: Wow that's so easy. Are there size limits on data uploaded to chatgpt?

Analyzing DynamoDB Data with ChatGPT-4 Advanced Data Analysis Instead of manually scanning your DynamoDB table for insights, take advantage of ChatGPT-4 Advanced Data Analysis to gain valuable insights for your app. For instance, if you want to visualize the number of accounts created on your app every week, follow these steps: 1. Export the data from your DynamoDB table as a CSV using the console. 2. Before uploading any data, anonymize customer information. Remove or replace fields like names and email addresses. Ensure that the account creation time is included in the dataset. 3. Upload the anonymized file to ChatGPT-4 and select the "Advanced Data Analysis" option. 4. Ask ChatGPT: "Can you plot a weekly graph of the accounts created?" That's it! You'll then have a visual representation of the accounts created on your app on a weekly basis.

R to @kgmodi: Emotional baggage

R to @Dj__: what does this mean?

If you are looking to find customers for your app after you have developed it, you will face an uphill battle. Instead, validate the demand before conceptualizing the idea. How? 1. Identify Keywords Related to Your Idea - let's say you want to create an app that automatically creates playlists based on taste, artist, and albums. - brainstorm keywords related to this idea. You can use tools like Ahrefs' keyword generator tool for assistance. - "music playlist," "playlist maker," "playlist generator," and "Spotify playlist" are some common keywords. 2. Assess the Demand for Each Keyword - ahrefs' keyword tool and Google Keyword Planner provide estimates for the search volume of these keywords. - e.g., "music playlist" has 2.4k searches per day, "playlist" has about 11k searches per day, and "Spotify playlist maker" has about 1k searches per day. 3. Determine Whether the Idea Is Worth Pursuing - a total volume of more than 1k searches indicates there is some demand for your new idea. - if your idea is applicable to more than one country, add the search volumes. If applicable, account for price parity in your markets. - in our example, "Spotify playlist maker" and "music playlist" easily surpass 1k in demand volume in the US. - from a demand perspective, the idea is worth pursuing. 4. Understand That Demand Is Only Half the Story - alongside demand, you should also formulate a plan to market and distribute your app before building it. - in our example, we need to determine how our customers will learn about our new "playlist app." - depending on your resources, plan your marketing and distribution strategy before building the app, not after. - common methods include Google ads, SEO and promotion through TikTok, Instagram, Newsletters and Twitter etc. - if you are going the SEO route, note that keyword difficulty plays a big role in getting the keyword ranked on the first page of the search results. To overcome this get a good domain name and plan for backlinks to your app. 5. Now, build your app.

R to @JustineCloudDev: Yes, I have created CDK templates to do some these. Let me fix them up so that I can share them.

R to @kgmodi: It's amazing what this guy has built at scale. Didn't even know something like that can be built in such a big place.