Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Build an intelligent voicemail system that transcribes messages, scores priority, extracts action items, and schedules callbacks automatically. Covers voicemail detection, message processing, and smart notifications.
Create an AI writing coach that provides layered feedback on grammar, style, structure, and tone — with actionable revision suggestions and progress tracking across writing sessions.
Master caching strategies for AI agents — from response caching and embedding caching to tool result caching and smart invalidation — to reduce latency, cut API costs, and improve throughput.
Design a multi-layer caching architecture for AI agent systems using Redis, application-level caches, and TTL strategies to reduce latency and LLM API costs while preventing cache stampedes and stale data problems.
Learn how to implement exact-match, semantic, and hybrid caching for AI agent responses. Achieve 30-60% cost reduction with proper cache architecture, hit rate optimization, and smart invalidation strategies.
Build an AI calendar agent that prepares meeting briefs, generates post-meeting summaries with action items, and sends automated follow-up emails using Google Calendar webhooks.
Build a complete call analytics pipeline that records calls, transcribes them, and extracts actionable insights using AI. Covers recording APIs, speaker diarization, sentiment analysis, and trend detection.
Master the three call transfer patterns for AI voice agents: cold transfer, warm transfer, and conferencing. Covers context passing, hold music, agent whisper, and seamless handoff implementation.
Implement canary deployments for AI agent systems with traffic splitting, health checking, automated rollback, and progressive delivery strategies that catch regressions before they affect all users.
