AI Developmentr/LocalLLaMA

I classified 3.5M US patents with Nemotron 9B on a single RTX 5090 — then built a free search engine on top

Read original
ai-coding-toolslocal-llm-workflowsdomain-specific-aihybrid-search-architecture

“Patent attorneys need exact phrase matching. 'solid-state battery electrolyte' should match those exact words, not semantically similar documents about 'energy storage.' FTS5 gives sub-second queries on 3.5M records with zero external dependencies.”

Key takeaways

  • Local LLM (Nemotron 9B on RTX 5090) classified 3.5M patents in 48 hours, demonstrating consumer hardware can handle enterprise-scale classification tasks
  • Contrarian architecture choice: FTS5 full-text search outperforms vector embeddings for domain-specific use cases requiring exact phrase matching and deterministic results
  • Complete technical stack disclosed: SQLite FTS5 + local LLM query expansion + BM25 ranking with custom weights + FastAPI, hosted on Chromebook via Cloudflare Tunnel - proving production-grade search doesn't require cloud infrastructure
  • Patent lawyer with 1 month coding experience built production search engine, signaling democratization of AI tooling for domain experts without traditional engineering backgrounds
  • Hybrid approach: Uses LLM for natural language query expansion into boolean queries, then traditional search for retrieval - combining strengths of both paradigms

Why this matters for operators: Legal tech, enterprise search architecture, local LLM deployment for specialized domains, hybrid search strategies

I cover AI×GTM intelligence like this every Wednesday.

Get STEEPWORKS Weekly

More picks

Personal Productivity & AI-Augmented WorkPractitioner StoryLenny's NewsletterVictor's pick

From Figma to Claude Code and back | Gui Seiz & Alex Kern (Figma)

Visual design one of last frontiers so this an interesting read

  • Figma's MCP enables bidirectional sync between design files and production code, eliminating static handoffs and version drift
  • AI coding tools enable 90% code generation when codebase is properly structured, with custom skills automating pre-flight checks and CI monitoring
  • Design workflow is bifurcating: AI handles rushed middle execution phase while humans focus on strategic planning (upstream) and quality craft (downstream)
ai-coding-toolscursor-vs-copilotautomation-stacks
Read original Full analysis →
AI×GTMThought LeadershipThe Signal (Brendan Short)Victor's pick

26 FAQs about GTM Engineering in 2026

A good if high level overview

  • Article is a FAQ format covering GTM Engineering predictions/practices for 2026
  • Published by Brendan Short's The Signal newsletter (6,595+ subscribers)
  • Content truncated - only intro/header visible, preventing full analysis
gtm-engineeringsignal-infrastructureenrichment
Read original Full analysis →
GTM OpsLenny's Newsletter

A guide to advanced B2B positioning

  • Content is a podcast/video episode announcement, not substantive article
  • Focuses on positioning fundamentals and cross-functional alignment challenges
  • No specific case studies, metrics, or implementation details provided in excerpt
back-to-basics-gtmvibe-marketing
Read original Full analysis →

This analysis was produced using the STEEPWORKS system — the same agents, skills, and knowledge architecture available in the GrowthOS package.