Kiernan

Kiernan's Substack focuses on the intersection of AI, technology development, and personal productivity. It examines the implications of AI advancements, habit formation, and the challenges of building and maintaining digital products. The author combines technical exploration with practical advice on navigating the modern digital landscape.

AI Development Personal Productivity Technology Innovation Digital Product Building Data Analysis Search Optimization Content Generation

The hottest Substack posts of Kiernan

And their main takeaways

Stealing fire from the gods

0 implied HN points • 06 Nov 23
🕹 Technology AI Data Content Generative AI
  1. Technological achievements have irreversible consequences that humanity must learn to live with.
  2. Generative AI is producing content so human-like that it's challenging to distinguish from real, impacting the way we process data.
  3. Data providence and trustworthiness will be crucial in a world flooded with low-quality generated content.

Starting with the Answer

0 implied HN points • 09 Sep 23
🕹 Technology Artificial Intelligence Search Machine Learning AI Applications
  1. Embedding vectors provide numerical representations for different types of content, allowing for easy comparison and search based on similarity.
  2. Starting with the answer in search means casting a broad net by providing an example of what you're looking for, rather than specific keywords.
  3. By utilizing embedding vectors, search results can be tailored to match concepts or sentiments, making searches more efficient and effective.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Relaunching Siev

0 implied HN points • 09 Aug 23
🕹 Technology Engineering Data insights
  1. Stabilizing the system by fixing shaky foundations for a more robust design.
  2. Relaunching Siev with new features like a cleaned up topic page, rich transcripts, and speaker identification page.
  3. Siev shaping up to be an advanced audio processing pipeline that can provide insights without needing to listen to entire streams.

Building in Front of Friends

0 implied HN points • 20 Apr 23
🕹 Technology AI Data Analysis Podcasts
  1. The author left their job at Clearbit after 5 years to launch into something new.
  2. The author is exploring AI and analyzing podcast data to extract valuable insights.
  3. Documentation of the author's ideas and projects is shared on their Substack, following a 'build in public' approach.

Building an ad detector the hard way

0 implied HN points • 12 May 23
🕹 Technology AI Data Analysis Transcription
  1. The ad detector is a work in progress, needing more refinement to distinguish ads from general content.
  2. The detector combines AI models to analyze show content and identify potential advertisements.
  3. Next steps involve improving accuracy, creating a web UI, and expanding the backlog of indexed audio content.

Accidentally building an enrichment product

0 implied HN points • 28 Apr 23
🕹 Technology Data Analysis Podcasts AI Tools
  1. Building an enrichment product unintentionally by combining different tools and systems.
  2. The product provides insights on ad placement in podcasts, changes in conversation, trends, and influencer sourcing.
  3. The project showcases the value of combining tools in a unique way and explores various potential use-cases.

Bridging the Gap

0 implied HN points • 05 May 23
🕹 Technology AI Data Analysis Machine Learning Podcasts
  1. The system can analyze podcast content like topics and sentiment without manual listening.
  2. Bridging the gap refers to improving machine trustworthiness for human tasks.
  3. Future plans involve deeper data analysis, such as identifying different types of ads in podcasts.

Another cog in the machine

0 implied HN points • 03 Jun 23
🕹 Technology Machine Learning NLP AI Data Analysis Software Development
  1. LLMs have limitations but can be powerful tools for specific tasks like identifying content in podcast transcripts.
  2. LLMs can be used to extract information from unstructured content, converting human-usable text into computer-usable formats with text instructions.
  3. Using LLMs for specific, constrained tasks can lead to quicker and more confident results compared to complex rule-based approaches.

Building an Speaker Identification Database

0 implied HN points • 14 Jul 23
🕹 Technology Data Analysis Transcription
  1. Creating a speaker identification database by utilizing existing data can be achievable in a short amount of time.
  2. Manually labeling missing speakers can enhance the accuracy and functionality of the database.
  3. Segmented transcripts based on speaker identification can enrich the overall user experience.