June 10, 2026• 8 min read
Why I Built OmniSLM
The AI tooling ecosystem is fragmented. Every local LLM project requires stitching together RAG, memory, and agents from scratch. OmniSLM is my attempt at a unified framework.
OmniSLMArchitecturePython
Engineering Notes
Technical deep dives, architecture decisions, and lessons learned while building production AI systems.
May 22, 2026• 12 min read
Building Local AI Systems
Enterprise data privacy requires local execution. Exploring the tradeoffs between Ollama, vLLM, and Llama.cpp for production workloads on edge hardware.
Local LLMPrivacyInfrastructure
April 15, 2026• 10 min read
Designing Multi-Tenant AI Applications
How to isolate vector databases, manage tenant-specific model configurations, and handle billing boundaries in Spring Boot applications.
Spring AISystem DesignJava
March 8, 2026• 7 min read
Lessons From Spring AI
Migrating enterprise Java teams to modern AI patterns. What Spring AI gets right, where it falls short compared to LangChain, and how to bridge the gaps.
Spring BootEnterprise Architecture
February 18, 2026• 15 min read
RAG Systems Beyond Tutorials
Standard dense retrieval isn't enough for production. Deep dive into hybrid search (FAISS + BM25), cross-encoder re-ranking, and dynamic chunking strategies.
RAGFAISSMachine Learning