Explore the Practical Side of ML & AI: Real-World Applications and Impact.https://ajing.github.io/https://ajing.github.io/posts/2026-05-28-agent-eval-difficulty-trajectory-constraints/https://ajing.github.io/posts/2026-05-28-agent-eval-difficulty-trajectory-constraints/A practical framework for making agent benchmarks harder in a controlled way: treat difficulty as trajectory-graph complexity, not prompt wording. Covers deterministic scoring, capability facets, harness effects, and systematic data generation.Thu, 28 May 2026 08:00:00 GMThttps://ajing.github.io/posts/2026-04-13-mercor-breach-what-leaked-about-ai-training/https://ajing.github.io/posts/2026-04-13-mercor-breach-what-leaked-about-ai-training/A $10B AI data vendor was breached, exposing 84 Airtable workspaces of training data for OpenAI, Anthropic, Apple, Amazon, and Meta. This post analyzes what the public reporting reveals about each lab's evaluation methodology — rubric design, RLHF pipelines, and quality control — and what it means for the industry.Mon, 13 Apr 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-03-11-improving-llm-i18n-tool-use-agency/https://ajing.github.io/posts/2026-03-11-improving-llm-i18n-tool-use-agency/LLMs achieve 57% tool-calling accuracy in English but only 34% across 52 languages — and 6.8% for the worst. This post covers the full playbook for closing the multilingual gap: training-time techniques, agentic architecture patterns, failure mode analysis, and RL-based approaches for i18n.Wed, 11 Mar 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-03-08-unverifiable-rewards-rl-frontier/https://ajing.github.io/posts/2026-03-08-unverifiable-rewards-rl-frontier/Deep research on tasks with unverifiable rewards in RL — the key bottleneck for scaling RL beyond math and code. Covers JEPO, NRT, RLNVR, self-play methods, GenRM, Constitutional AI, reward hacking mitigation, and more.Sun, 08 Mar 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-03-05-instruction-following-post-training-data/https://ajing.github.io/posts/2026-03-05-instruction-following-post-training-data/LLMs can write poetry and solve math, but ask them to 'respond in exactly 3 bullet points using only lowercase' and they stumble. This post dissects the taxonomy of instruction-following failures and provides a practical playbook for building post-training data that actually fixes them.Thu, 05 Mar 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-03-01-experience-augmented-icl-complement-to-rl-post-training/https://ajing.github.io/posts/2026-03-01-experience-augmented-icl-complement-to-rl-post-training/RL post-training makes models smarter, but it can't cover the infinite long tail of real-world cases. Experience-augmented ICL retrieves successful reasoning traces at inference time, letting agents learn continuously from real usage — no retraining required.Sun, 01 Mar 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-01-10-tool-selection-optimization-llm-agents-at-scale/https://ajing.github.io/posts/2026-01-10-tool-selection-optimization-llm-agents-at-scale/A deep technical dive into tool selection—retrieval strategies, context optimization, learned selection, and the engineering trade-offs that matter when scaling to hundreds of tools.Sat, 10 Jan 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-01-03-generative-engine-optimization-geo/https://ajing.github.io/posts/2026-01-03-generative-engine-optimization-geo/A comprehensive guide to Generative Engine Optimization—making your content retrievable, citable, and recommendable by large language models.Sat, 03 Jan 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-01-02-ads-formats-in-llm-products/https://ajing.github.io/posts/2026-01-02-ads-formats-in-llm-products/A survey of advertising formats in LLM products—separating what's deployed in production from what remains in research.Fri, 02 Jan 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-01-02-ads-in-llm-chatbot/https://ajing.github.io/posts/2026-01-02-ads-in-llm-chatbot/Exploring how to integrate ads in LLMs through character training—making recommendations genuinely helpful rather than annoyingly promotional.Fri, 02 Jan 2026 00:00:00 GMThttps://ajing.github.io/posts/2026-01-01-generative-ui-doesnt-move-the-needle-steering-does/https://ajing.github.io/posts/2026-01-01-generative-ui-doesnt-move-the-needle-steering-does/After shipping multiple generative UI features, I've concluded that the sophistication of AI-generated interfaces often doesn't translate to user benefit—but steering does.Thu, 01 Jan 2026 00:00:00 GMThttps://ajing.github.io/posts/2025-12-31-rlhf-engineering-implementation/https://ajing.github.io/posts/2025-12-31-rlhf-engineering-implementation/A practical engineering guide to RLHF implementation—covering PPO, GRPO, DPO, and tool-use training with code snippets and debugging tips.Wed, 31 Dec 2025 00:00:00 GMThttps://ajing.github.io/posts/2025-12-31-rlhf-ppo-dpo-grpo-notes/https://ajing.github.io/posts/2025-12-31-rlhf-ppo-dpo-grpo-notes/Reading notes on RLHF covering PPO, DPO, and GRPO—understanding post-training as an engineering pipeline rather than a single algorithm.Wed, 31 Dec 2025 00:00:00 GMThttps://ajing.github.io/posts/2025-09-07-ae-vae-training-learnings/https://ajing.github.io/posts/2025-09-07-ae-vae-training-learnings/Practical lessons from training autoencoders and VAEs for embedding compression—covering dimension choice, KL scheduling, contrastive signals, and evaluation metrics.Sun, 07 Sep 2025 00:00:00 GMThttps://ajing.github.io/posts/2025-09-07-user-interest-modeling-with-transformer-architectures/https://ajing.github.io/posts/2025-09-07-user-interest-modeling-with-transformer-architectures/Exploring position embeddings, architecture choices, and training techniques for Transformer-based recommender systems.Sun, 07 Sep 2025 00:00:00 GMThttps://ajing.github.io/posts/2025-05-03-ui-representation-action-execution/https://ajing.github.io/posts/2025-05-03-ui-representation-action-execution/Exploring structured UI representation using JSON Schema, and how to implement action handlers for generative UI systems.Sat, 03 May 2025 00:00:00 GMThttps://ajing.github.io/posts/2025-04-01-generative-ui/https://ajing.github.io/posts/2025-04-01-generative-ui/Exploring how LLMs can dynamically generate user interfaces that adapt in real time to user needs—the vision behind generative UI.Tue, 01 Apr 2025 00:00:00 GMThttps://ajing.github.io/posts/2023-08-20-how-to-make-llm-serving-faster/https://ajing.github.io/posts/2023-08-20-how-to-make-llm-serving-faster/An overview of LLM inference optimization techniques including KV cache, FlashAttention, and memory management strategies.Sun, 20 Aug 2023 00:00:00 GMT