https://jnk234.github.io/posts/sycophancy-recovery-simpo/2026-04-01T00:00:00+00:00https://jnk234.github.io/2026-04-01T00:00:00+00:00https://jnk234.github.io/posts/2026-04-01T00:00:00+00:00https://jnk234.github.io/posts/sycophancy-recovery-dpo/2026-03-29T00:00:00+00:00https://jnk234.github.io/posts/deepseek-mhc-hyper-connections/2026-01-23T00:00:00+00:00https://jnk234.github.io/posts/understanding-rmsnorm-my-notes-on-faster-layer-normalization/2026-01-03T00:00:00+00:00https://jnk234.github.io/projects/2025-12-20T00:00:00+00:00https://jnk234.github.io/projects/quibo-mcp-server-for-agentic-blogging/2025-12-20T00:00:00+00:00https://jnk234.github.io/projects/mcp-multiverse/2025-12-18T00:00:00+00:00https://jnk234.github.io/projects/vouchai-agent-to-agent-insurance-protocol/2025-11-15T00:00:00+00:00https://jnk234.github.io/projects/self-evolving-agent/2025-11-04T00:00:00+00:00https://jnk234.github.io/posts/the-deadly-triad-in-reinforcement-learning-why-agents-fail-and-how-dqn-fixed-it/2025-09-22T00:00:00+00:00https://jnk234.github.io/projects/reinforcement-learning-algorithms/2025-09-21T00:00:00+00:00https://jnk234.github.io/posts/a-deep-dive-into-q-learning-the-off-policy-td-control-algorithm/2025-09-19T00:00:00+00:00https://jnk234.github.io/posts/a-deep-dive-into-on-policy-td-control-the-sarsa-algorithm/2025-09-18T00:00:00+00:00https://jnk234.github.io/posts/temporal-difference-bootstrapping-in-reinforcement-learning/2025-09-17T00:00:00+00:00https://jnk234.github.io/posts/monte-carlo-learning-in-rl/2025-09-15T00:00:00+00:00https://jnk234.github.io/posts/model-free-rl-prediction-control-and-the-mrp-mdp-duality/2025-09-14T00:00:00+00:00https://jnk234.github.io/posts/my-three-months-at-relativity-building-ai-for-legal-tech/2025-09-07T00:00:00+00:00https://jnk234.github.io/projects/nlp-interpretability-mechanistic-analysis-of-llms/2025-09-01T00:00:00+00:00https://jnk234.github.io/posts/reinforcement-learning-essentials-mdps-optimal-control/2025-08-09T00:00:00+00:00https://jnk234.github.io/posts/beyond-supervised-learning-unlocking-ais-potential-with-reinforcement-learning/2025-07-14T00:00:00+00:00https://jnk234.github.io/projects/lear-llm-driven-evolution-of-agent-based-rules/2025-07-01T00:00:00+00:00https://jnk234.github.io/projects/netlogo-llm-extension/2025-06-19T00:00:00+00:00https://jnk234.github.io/projects/air-insights-legal-document-intelligence/2025-06-01T00:00:00+00:00https://jnk234.github.io/projects/qd-lear-quality-diversity-in-llm-evolved-agents/2025-05-01T00:00:00+00:00https://jnk234.github.io/projects/medhastra-ai-medical-education-platform/2025-03-22T00:00:00+00:00https://jnk234.github.io/projects/faceswap-diffusion-model/2025-03-20T00:00:00+00:00https://jnk234.github.io/projects/advocate-ai-powered-ad-generator/2025-03-12T00:00:00+00:00https://jnk234.github.io/projects/second-opinaion-medical-diagnosis-system/2025-03-12T00:00:00+00:00https://jnk234.github.io/projects/agentic-blogging-assistant/2025-02-18T00:00:00+00:00https://jnk234.github.io/posts/implementing-gpt-style-attention-a-step-by-step-guide-with-pytorch/2025-01-21T00:00:00+00:00https://jnk234.github.io/posts/the-ultimate-guide-to-preparing-text-data-for-language-modeling-with-pytorch/2025-01-06T00:00:00+00:00https://jnk234.github.io/posts/cdatapytorch-in-practice-essential-building-blocks-for-modern-deep-learning/2025-01-02T00:00:00+00:00https://jnk234.github.io/posts/pytorch-in-practice-essential-building-blocks-for-modern-deep-learning/2025-01-02T00:00:00+00:00https://jnk234.github.io/posts/pytorch-in-practice-essential-building-blocks-for-modern-deep-learning/2024-12-30T00:00:00+00:00https://jnk234.github.io/posts/value-based-policy-training-in-reinforcement-learning/2024-09-30T00:00:00+00:00https://jnk234.github.io/posts/understanding-reinforcement-learning-policy-based-and-value-based-approaches/2024-09-02T00:00:00+00:00https://jnk234.github.io/posts/fast-and-efficient-finetuning-of-llms-qlora/2024-08-19T00:00:00+00:00https://jnk234.github.io/posts/from-decisions-to-rewards-understanding-the-rl-decision-making-process/2024-08-14T00:00:00+00:00https://jnk234.github.io/posts/reinforcement-learning-essentials-a-quick-guide/2024-08-12T00:00:00+00:00https://jnk234.github.io/posts/learn-act-adapt-unveiling-reinforcement-learning/2024-08-05T00:00:00+00:00https://jnk234.github.io/posts/from-perplexity-to-rouge-essential-metrics-for-evaluating-llms-easy-to-understand/2024-04-08T00:00:00+00:00https://jnk234.github.io/posts/can-less-be-more-exploring-peft-for-llms/2024-03-27T00:00:00+00:00https://jnk234.github.io/posts/decoding-the-art-understanding-text-generation-with-transformers-ii/2024-03-26T00:00:00+00:00https://jnk234.github.io/posts/decoding-the-art-understanding-text-generation-with-transformers-i/2024-03-25T00:00:00+00:00https://jnk234.github.io/posts/unlock-the-power-of-generative-ai-mastering-personalized-model-development/2024-01-22T00:00:00+00:00https://jnk234.github.io/projects/neuraforge-newsletter/2023-08-01T00:00:00+00:00https://jnk234.github.io/projects/technical-publications-automation/2022-07-01T00:00:00+00:00https://jnk234.github.io/posts/cdataheres-how-you-should-train-an-intelligent-classifier-model../2021-12-02T00:00:00+00:00https://jnk234.github.io/posts/heres-how-you-should-train-an-intelligent-classifier-model../2021-12-02T00:00:00+00:00https://jnk234.github.io/posts/cdatawhy-multi-label-classification-should-be-used-instead-of-conventional-classifiers./2021-11-29T00:00:00+00:00https://jnk234.github.io/posts/why-multi-label-classification-should-be-used-instead-of-conventional-classifiers./2021-11-29T00:00:00+00:00https://jnk234.github.io/posts/cdataapproaching-data-centric-ai-using-fast.ai/2021-11-06T00:00:00+00:00https://jnk234.github.io/posts/approaching-data-centric-ai-using-fast.ai/2021-11-06T00:00:00+00:00https://jnk234.github.io/posts/cdatagetting-started-with-100-days-of-deep-learning/2021-10-20T00:00:00+00:00https://jnk234.github.io/posts/getting-started-with-100-days-of-deep-learning/2021-10-20T00:00:00+00:00https://jnk234.github.io/cv/https://jnk234.github.io/publications/