Best of Daily Dose of Data Science | Avi Chawla | Substack — January 2026

1. 1
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·11w

   [New] Generative UI for Agents

   Generative UI is an emerging pattern where AI agents render actual UI components instead of just returning text responses. Unlike traditional chat interfaces, agents can now display weather cards, confirmation dialogs, data tables, and other interactive elements by selecting pre-built components and filling them with data at runtime. Three approaches exist: static (predefined components), declarative (component registry), and open-ended (raw HTML/iframes). Protocols like A2UI, AG-UI, and MCP Apps enable real-time bidirectional communication between agents and frontends. CopilotKit has open-sourced a complete implementation for React with integrations for LangGraph, CrewAI, and other agent frameworks. MiniMax also launched Agent Desktop, a desktop environment where AI agents can browse the web, manage files, and automate developer tasks.

   109

   4
2. 2
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·11w

   Phases of ML Modeling

   ML systems should evolve through four distinct phases rather than jumping straight to complex models. Start with simple heuristics and rules (Phase 1), then move to basic ML models like logistic regression (Phase 2), optimize through feature engineering and hyperparameter tuning (Phase 3), and only adopt complex models like deep neural networks when simpler approaches are exhausted (Phase 4). This staged approach reduces risk, improves debuggability, and ensures each phase's best model becomes the baseline for the next, encouraging incremental progress and evidence-driven decision-making.

   90

   1
3. 3
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·13w

   6 Components of Context Engineering

   Context engineering is the practice of optimizing how information flows to AI models, comprising six core components: prompting techniques (few-shot, chain-of-thought), query augmentation (rewriting, expansion, decomposition), long-term memory (vector/graph databases for episodic, semantic, and procedural memory), short-term memory (conversation history management), knowledge base retrieval (RAG pipelines with pre-retrieval, retrieval, and augmentation layers), and tools/agents (single and multi-agent architectures, MCPs). While model selection and prompts contribute only 25% to output quality, the remaining 75% comes from properly engineering these context components to deliver the right information at the right time in the right format.

   87

   3
4. 4
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·12w

   Your RAG System Has a Hidden UX Problem

   RAG systems often use semantic retrieval but fall back to keyword-based highlighting, creating a UX disconnect where users can't see why documents are relevant. Zilliz released an open-source semantic highlighting model that identifies semantically relevant text spans instead of just keyword matches. The bilingual model (English/Chinese) handles 8K context windows, runs fast enough for production use, and outperforms existing solutions on both in-domain and out-of-domain benchmarks. It's being integrated into Milvus as a native API and is available on Hugging Face under MIT license.

   26
5. 5
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·14w

   Foundations of AI Engineering and LLMOps

   Part 3 of an LLMOps course is now available, covering attention mechanisms, transformer architectures, mixture-of-experts, and the fundamentals of pretraining and fine-tuning with hands-on code demos. LLMOps extends traditional MLOps principles to address the unique engineering challenges of managing large language models like Llama, GPT, and Claude in production, focusing on reliability, accuracy, security, and cost-effectiveness. The course aims to provide systems-level thinking for building production-ready LLM applications with clear explanations, examples, diagrams, and implementations.

   25
6. 6
   Article

   Daily Dose of Data Science | Avi Chawla | Substack·11w

   Build Agents That Can Learn Like Humans

   ART (Agent Reinforcement Trainer) is an open-source framework that simplifies reinforcement learning for LLMs by eliminating manual reward function engineering. It uses GRPO (Group Relative Policy Optimization) where agents attempt tasks multiple times, an LLM judge compares attempts, and the model learns from relative performance. Unlike traditional RL frameworks limited to simple chatbot interactions, ART supports multi-turn conversations, tool calls, and integrates with LangGraph, CrewAI, and ADK. It combines vLLM for model serving and Unsloth for GRPO training, enabling developers to fine-tune small open-source models to outperform larger closed-source alternatives on specific tasks.

   14