<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Llm on Aditya Singh</title><link>https://singhaditya8499.github.io/tags/llm/</link><description>Recent content in Llm on Aditya Singh</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 30 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://singhaditya8499.github.io/tags/llm/index.xml" rel="self" type="application/rss+xml"/><item><title>Dialectic: Two-LLM Debate Studio</title><link>https://singhaditya8499.github.io/projects/dialectic/</link><pubDate>Mon, 30 Mar 2026 00:00:00 +0000</pubDate><guid>https://singhaditya8499.github.io/projects/dialectic/</guid><description>Debate-first AI system where two models argue opposite sides with evidence anchors, live streaming transcript, and structured high-fidelity summaries.</description></item><item><title>CardWise AI</title><link>https://singhaditya8499.github.io/projects/cardwise-ai/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://singhaditya8499.github.io/projects/cardwise-ai/</guid><description>AI platform for credit-card discovery, financial planning, and spending summaries.</description></item><item><title>DocCompass: Local-First Document Search Agent</title><link>https://singhaditya8499.github.io/projects/doccompass/</link><pubDate>Sat, 21 Mar 2026 00:00:00 +0000</pubDate><guid>https://singhaditya8499.github.io/projects/doccompass/</guid><description>Semantic local document search system with lazy indexing, Ollama-powered query planning, fallback heuristics, and scored ranked results.</description></item><item><title>Improving LLM Alignment in Conversational Analytics</title><link>https://singhaditya8499.github.io/posts/improving-llm-alignment/</link><pubDate>Mon, 05 Jan 2026 00:00:00 +0000</pubDate><guid>https://singhaditya8499.github.io/posts/improving-llm-alignment/</guid><description>Prompt and retrieval strategies, evaluation loops, and safety checks for production readiness.</description></item><item><title>GPT Stock Recommender</title><link>https://singhaditya8499.github.io/projects/gpt-stock-recommender/</link><pubDate>Thu, 01 Jan 2026 00:00:00 +0000</pubDate><guid>https://singhaditya8499.github.io/projects/gpt-stock-recommender/</guid><description>AI-driven stock recommendation pipeline with sentiment-aware evaluation.</description></item></channel></rss>