Sun, Dec 15, 2024

Weekly Notes and Discoveries | S Anand

**/*.md can search for all Markdown files. Julia Evans #markdown

Windows 11 2024 Update features: Ref
- Live captions (via the tray) can transcribe audio and microphone.
- Cocreator in Paint lets you draw crudely and enhances it with AI. The neat UI is a slider that lets you control how close it should be to your drawing.
- Voice Clarity automatically cancels echo, reduces background noise, and minimizes reverb.
- Studio Effects (via the tray) lets you apply camera effects on all apps. Eye contact feature is CLEVER!
- sudo lets you run commands with admin privileges from the command line. source

Traffic to StackOverflow has fallen considerably. Especially from young and Indian developers. StackOverflow revenue is down. Via Prashanth. They're exploring:
- Licensing their content. (Meta says high quality content improves LLM performance by 30% on HumanEval)
- Enterprise StackOverflow for system integration
- Fine-tuned versions of Enterprise Stackoverflow for enterprises
- Integrate StackOverflow within your IDE. Ask questions, post directly

Hyperbrowser is a cloud based puppeteer service. #cloud #future

Hacker News RSS is a good way to get RSS feeds from Hacker News. It's also a good way to understand how to convert a news source into RSS feeds. BlueSky has RSS feeds too

Roaming RAG is an alternative to RAG without the vector database. #markdown
- Applicable to well structured documents, e.g. technical books, manuals, etc.
- Create a hierarchical outline of the document. Code
  - Keep the top-level headings.
  - Preserve the first ~100 characters of opening text from each section.
  - Present the second-level headings, but without any subsidiary content.
  - Provide each section a unique 8 digit hex identifier.
  - Each section heading is followed by a guiding comment for the model: Section collapsed - expand with expand_section("{identifier}").
- Then read the relevant sections as context to answer the question. Code

I surveyed the Gramener QA team on how they were using LLMs. #llm-ops
- 7 used it for code generation (e.g. date extraction, regex generation)
- 4 used it for learning (e.g. Robot Framework, how to define test cases, API usage)
- 3 used it for formula generation (e.g. Excel)
- 2 used it for test scenario identification
- 2 used it for test data generation
- 2 used it for comparing expected vs actual datasets
- 1 used it for data type identification (e.g. given sample values, identify the data type).
- 1 used it for evaluating resulting (LLM as a judge)

I asked the Straive Digitalized Operations team what management techniques they would apply to manage LLMs. Here are the responses: #automation #best-practices #llm-ops #optimization #prompt-engineering
- Ask better questions. (Prompt engineering.)
- Create templates or step-by-step instructions. (Chain of Thought.)
- Ask for multiple options and pick from the best options. (Agentic approach?)
- Training. (Fine tuning.)
- Price weaker responses lower. (Stratified model pricing?)

"LLM hallucinations are a good thing. They are a sign of diversity, allowing us to improve the answer by exploring multiple paths." -- A colleague from Straive. #future #llm-ops

Bedrock Llama models can't be directly called with their model names. You need to use their inference profile names, e.g. us.meta.llama3-2-11b-instruct-v1:0 if the model is in a US region. #llm-ops

When embedding using a SentenceTransformer.encode(docs) it's best if we embed with smaller docs and call it multiple times (rather than embedding more at once). On Colab T4, for gte-base-en-v1.5, when embedding 1,000 docs of up to 8K chars each, here is the TOTAL time it took, based on batch sizes (lower is better) #embeddings
- 1 doc per call: 10s
- 2 docs per call: 13s
- 4 docs per call: 19s
- 8 docs per call: 23s
- 16 docs per call: 32s
- 32 docs per call: 40s

Running embeddings without a GPU is extremely slow. It takes ~2.4 seconds per string. #embeddings #gpu