Posts

[Feb 25] Interesting Things I Learnt This Week

Image
1. How a group of birds got their names - This is an article about the collective names of groups of birds. It discusses how these names came to be and how they have changed over time. The names are often based on the birds’ characteristics or behavior. Some examples are a “murder of crows” and a “parliament of owls”. The article also mentions a book called “ The Boke of Seynt Albans ” which contains a list of many of these collective names. My Take: The evolution of names across centuries is a fascinating journey, often revealing a captivating rabbit hole of linguistic transformations. At times, it strikes me that we may become overly fixated on specific details that are destined to fade with the passage of time. In doing so, we risk losing sight of the broader picture. This phenomenon is particularly evident when exploring historical instances that highlight the intellectual contributions of women. Unfortunately, history illustrates how their brilliance was often stifled through fo

[Feb 18] - Interesting things I learnt this week

A few interesting things I learnt about this week are 1. Coffee Stains in LaTeX - This package provides an essential feature that LaTeX has been missing for too long: It adds coffee stains to your documents. A lot of time can be saved by printing stains directly on the page rather than adding them manually. My Take: There is too much fun to be had with LaTeX. I guess the rest of the world missed this wonderful piece of tech completely. These small and sometimes pointless tools are what really make any technology platform stand out among others, not because of the tech but the fact that its users are emotionally invested to build these kind of tech nuggets 2. Meshtastic - Meshtastic is a project that enables you to use inexpensive LoRa radios as a long range off-grid communication platform in areas without existing or reliable communications infrastructure. This project is 100% community driven and open source. LoRa uses license-free sub-gigahertz radio frequency bands EU868 (863–87

[Feb 10] - Interesting things I learnt this week

This is second of my experiment series to compile a few interesting things which I read about during the week. I hope I can continue this for this year at least. 1. Eagle 7B   - The race for better models is on. Gemini from Google is there but its another closed source one. But this one is open source. It claims to have transformers with 1 Trillion Tokens Across 100+ Languages.  My Take: Models are going to be get better in the open source world with each passing day. Very soon we will have GPT4 level models running on our local laptops and phones. The open source is going to be a big part of it. How people build on top of these models is going to become even more interesting. I have not tried this model yet, but I am guessing it will be faster than llama2  2. Text Generation WebUI - This is an amazing tool which starts a webservice on your machines and makes loading and configuring various LLMs a breeze. It will autodownload models once you have given hugginface project name.  My T

[Jan 28] - Interesting things I learnt this week

 This is my experiment to compile a few interesting things which I read about during the week. I hope I can continue this for this year at least.   1. One Pedal Driving : This is an article about one-pedal driving in electric cars. It discusses what one-pedal driving is and the benefits of using it. One-pedal driving allows you to accelerate and brake using only the gas pedal. This can help extend the life of your brakes and increase your car’s range. It also reduces stress while driving. The article also details how one-pedal driving works and how to use it safely.  My Take: This is really nice idea. I drive a smart hybrid which I kind of drive in a one pedal mode. Not sure if I am getting most bang for the buck, but I kind of am used to leaving the accelerator and the car slowing down and coming to a stop where I want (obviously I need to brake for complete stop) ever since I learnt driving first, thanks for my trainer.    2. Reading QR Codes manually - This guide teaches you

Lessons from Wonka, LLM Edition

 Being bored right after watching Wonka, I decided to try this out. Gave ChatGPT3.5, Bard, Mistral and llama2 the following prompt.  write a note about entrepreneurial lessons to be learnt from the story of wonka Following is what they generated  ChatGPT3.5 Subject: Entrepreneurial Lessons from the Story of Wonka Dear Entrepreneurs and Dreamers, In the whimsical world of Willy Wonka, where chocolate rivers flow and everlasting gobstoppers exist, there lies a treasure trove of entrepreneurial lessons waiting to be unwrapped. The tale of Willy Wonka, the eccentric chocolatier, is more than just a fantastical story; it's a guidebook for those daring enough to dream big and embark on their entrepreneurial journey. Unbridled Creativity: Willy Wonka's factory is a testament to the power of imagination. Entrepreneurs can learn that creativity knows no bounds. Innovate, think outside the box, and let your imagination run wild. Sometimes, the most extraordinary ideas come from the seem

Some thoughts on LLMs

The comparison between the current stage of Large Language Models (LLMs) and the introduction of databases in the computer industry is related to the potential impact of LLMs on various applications and industries. Just as databases provided a foundation for various computer systems like Google Search and Gmail, LLMs are becoming a foundation for various language-related applications. Companies like Google have already demonstrated promising results with multiple LLMs, such as Bard, at their I/O conference. OpenAI has also partnered with Microsoft to develop innovative LLM solutions. In addition, the open-source community is making significant contributions to the development of LLMs, as seen with projects like Vicuna. Looking into the future, the open-source solutions may play an essential role in running custom-tuned on-premise or on-device LLMs. This highlights the potential for innovative solutions and applications that can be built on top of these LLMs. One industry that could ben

OKRs and SMART Goals : When and Why they fail

Image
OKR, or Objectives and Key Results, is a method for setting and tracking progress toward specific, measurable, and time-bound goals. The framework consists of two parts: objectives, which are the high-level aspirations an organization or individual wants to achieve, and key results, which are the specific and measurable steps needed to reach each objective. OKR was created by former Intel CEO Andy Grove and has been adopted by companies like Google, LinkedIn, and Twitter. The purpose of the OKR system is to align goals, focus efforts, and measure progress towards objectives. SMART goals are objectives that are specific, measurable, attainable, relevant, and time-bound. The acronym was coined in 1981 by George Doran and is widely used as a framework for setting and achieving goals. A SMART goal is clear, quantifiable, achievable, aligned with broader goals, and has a deadline. This framework can help individuals and organizations to focus their efforts, track progress, and achieve their