Personal Reflections & Tech Insights

Blogs

Orchestrating LLM Workflows in Serverless

Building real-world LLM applications often means chaining multiple prompts, conditional logic, and retries. Serverless functions need orchestration to manage this state and complexity.

Scaling Reinforcement Learning in Production

Moving RL agents from research to production brings unique challenges. It's not just about the model, but the entire system around it.

Taming AI Microservices with a Service Mesh

AI workloads bring new complexity to microservices. A service mesh can help manage traffic, observability, and security for these demanding systems.

Catching Silent Failures in ML Pipelines

Your ML pipeline might be green, but your model could be failing silently in production due to data issues. Data observability helps you catch these problems.

The Catch with Computing on Encrypted Data

Homomorphic encryption lets you compute on data without decrypting it, a powerful idea for secure AI. But its complexity and performance hit are significant.

Why Your Production AI Needs Explanations

Deploying black-box AI models without understanding their decisions can lead to serious issues. Explainable AI (XAI) is critical for debugging, trust, and compliance.

Monitoring AI in production: Beyond the usual metrics

Observing AI/ML systems in production goes beyond traditional infrastructure metrics. It needs deep insights into data quality, model behavior, and performance to catch silent failures.

Kubernetes for ML: Orchestrating Production Models

Deploying machine learning models to production brings unique challenges. Kubernetes offers powerful tools for managing these complex workflows, but it's not a silver bullet.

Optimizing GPU Costs for Multi-Tenant AI Inference

Sharing GPUs for AI inference across multiple users or services is tricky. This post explores how to allocate these expensive resources efficiently without sacrificing performance or breaking the bank.