Start Here
A short map of the main themes on this site.
This site is about the practical side of ML and AI systems: how models, agents, training pipelines, evaluations, and product interfaces behave when they leave the paper and enter a real engineering loop.
The fastest entry point is to pick a path below, then follow the linked topic hub when you want the full archive.
Reading Paths
LLM Agents
Tool use, agent runtime design, evaluation, context, and production patterns for systems that act across tools and environments.
- Reproducing CompactRL: What Worked, What Failed, and Why We Did Not Scale
An auditable CompactRL reproduction spanning the public algorithm, a 96-step long-horizon simulation, integration with slime, real Qwen actor-critic training, value-function fixes, 17 experimental phases, and the evidence that stopped us from scaling.
- From Long CoT to Agent Swarms: The Documented Evolution of Kimi's Reinforcement Learning
A source-grounded history of Kimi's reinforcement-learning stack, from Kimi k1.5's long-context outcome RL and partial rollouts to K2's general RL and K2.5's multimodal GRMs and Parallel-Agent RL.
- Training the Critic Without Crashing the Reward: A Practical Guide to Agentic RL
A practical framework for critic training and credit assignment in long-horizon LLM agents: IQL, pairwise advantage, hindsight and counterfactual critics, privileged information, turn-level MDPs, chain-of-thought monitoring, and reward-crash diagnosis.
Evaluation
Practical approaches to measuring model and agent capability with deterministic checks, rubrics, trajectories, and verifiable outcomes.
- Reproducing CompactRL: What Worked, What Failed, and Why We Did Not Scale
An auditable CompactRL reproduction spanning the public algorithm, a 96-step long-horizon simulation, integration with slime, real Qwen actor-critic training, value-function fixes, 17 experimental phases, and the evidence that stopped us from scaling.
- From Long CoT to Agent Swarms: The Documented Evolution of Kimi's Reinforcement Learning
A source-grounded history of Kimi's reinforcement-learning stack, from Kimi k1.5's long-context outcome RL and partial rollouts to K2's general RL and K2.5's multimodal GRMs and Parallel-Agent RL.
- Training the Critic Without Crashing the Reward: A Practical Guide to Agentic RL
A practical framework for critic training and credit assignment in long-horizon LLM agents: IQL, pairwise advantage, hindsight and counterfactual critics, privileged information, turn-level MDPs, chain-of-thought monitoring, and reward-crash diagnosis.
Post-Training
SFT, RLHF, preference optimization, instruction following, reasoning traces, and data pipelines for shaping model behavior after pretraining.
- Reproducing CompactRL: What Worked, What Failed, and Why We Did Not Scale
An auditable CompactRL reproduction spanning the public algorithm, a 96-step long-horizon simulation, integration with slime, real Qwen actor-critic training, value-function fixes, 17 experimental phases, and the evidence that stopped us from scaling.
- From Long CoT to Agent Swarms: The Documented Evolution of Kimi's Reinforcement Learning
A source-grounded history of Kimi's reinforcement-learning stack, from Kimi k1.5's long-context outcome RL and partial rollouts to K2's general RL and K2.5's multimodal GRMs and Parallel-Agent RL.
- Training the Critic Without Crashing the Reward: A Practical Guide to Agentic RL
A practical framework for critic training and credit assignment in long-horizon LLM agents: IQL, pairwise advantage, hindsight and counterfactual critics, privileged information, turn-level MDPs, chain-of-thought monitoring, and reward-crash diagnosis.
RLHF and Preference Optimization
Engineering notes and research synthesis on PPO, DPO, GRPO, reward modeling, preference data, and model behavior optimization.
- Reproducing CompactRL: What Worked, What Failed, and Why We Did Not Scale
An auditable CompactRL reproduction spanning the public algorithm, a 96-step long-horizon simulation, integration with slime, real Qwen actor-critic training, value-function fixes, 17 experimental phases, and the evidence that stopped us from scaling.
- From Long CoT to Agent Swarms: The Documented Evolution of Kimi's Reinforcement Learning
A source-grounded history of Kimi's reinforcement-learning stack, from Kimi k1.5's long-context outcome RL and partial rollouts to K2's general RL and K2.5's multimodal GRMs and Parallel-Agent RL.
- Training the Critic Without Crashing the Reward: A Practical Guide to Agentic RL
A practical framework for critic training and credit assignment in long-horizon LLM agents: IQL, pairwise advantage, hindsight and counterfactual critics, privileged information, turn-level MDPs, chain-of-thought monitoring, and reward-crash diagnosis.
Generative UI
How AI systems can produce, steer, and execute user interfaces with structured representations and practical product constraints.
- Ad Formats in LLM Products: What's Live vs. What's Research
A survey of advertising formats in LLM products—separating what's deployed in production from what remains in research.
- Generative UI Doesn't 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.
- UI Representation and Action Execution for Generative UI
Exploring structured UI representation using JSON Schema, and how to implement action handlers for generative UI systems.