AI Lab
Grounded in production experience building AI agents on AWS Bedrock. Hands-on tutorials here, with shorter notes published on LinkedIn.
Tutorial series
View on GitHub ↗A hands-on series on building with LangChain and LangGraph — one lesson per pattern, from foundational prompt chaining to multi-agent systems and production observability. Each lesson pairs a deep-dive concept doc with a runnable notebook.
Sequential LLM calls with LCEL
Concurrent chains with RunnableParallel
Rule-based and semantic routing
Plan, delegate to workers, then synthesize
Generate, score, and refine in a loop
Let the model choose tools and drive its own loop
Thread memory vs. facts that outlive it
Window, trim, and summarize a growing transcript
Retrieve your data into the prompt, then upgrade it
Specialists, supervisors, and subgraphs
Feel fast with streaming, be fast with async
See every step and measure output quality
Roadmap
· 0 plannedWriting · LinkedIn
Post · LinkedIn
Aurora is a personal AI that runs entirely on your Android phone — offline, private, and free, with nothing ever leaving the device. Opening a small closed Android beta ahead of the Play Store launch.
Article · LinkedIn
The next leap in AI may come from efficiency, not scale — Google's TurboQuant shows aggressive vector compression can hold model quality while slashing memory and infra cost.
Also exploring