Solution Development with Local Language Models
- Building solutions based on local LLMs for business tasks: meeting and document summarization, RAG search across corporate knowledge bases, text data analysis.
- Architecture design, model selection, inference setup, and integration with existing business processes.
Subagents
- Legal RAG Navigator Searches agreements and regulatory documents, with quotations and links to sources.
- Automation with Obsidian and hybrid search Automatic meeting transcription: speaker identification, translation, summary, minutes, and an action list from a single recording.
- Amail — a business email assistant Locally classifies emails, explains priorities, finds corporate context, and prepares draft replies.
- Email sorting and draft response generation Consolidates scattered requirements into a coherent specification and acceptance criteria.
- Database queries via Text-to-SQL XLSX, CSV, and CRM exports are consolidated in PostgreSQL; users ask questions about the data by text or voice.
Agents
- CustDev Agent Turns interviews and notes into a map of observations, needs, and testable product opportunities.
- Muse Agent Checks the interface, user flow, and copy for clarity, accessibility, consistency, and honesty.
- Super Oracle Analyzes a product task using 5 methods (Fishbone, TRIZ, Blue Ocean, Pre-mortem, and RICE) and returns 5 artifacts and a plan.
- 2024 – present AI Solutions Developer Project work
