Topics

Building the models everything else runs on — LLM and multimodal architecture, pre-training, scaling laws, and evaluation.

• large language models
• LLMs
• transformers
• model architecture
• pre-training
• scaling laws
• +11

Putting AI to work in real products — RAG, agents, fine-tuning, prompt and context engineering, tool use, and vector search.

• RAG
• retrieval augmented generation
• AI agents
• agentic workflows
• embeddings
• fine-tuning
• +17

Making models run fast, cheap, and at scale — GPUs, inference optimization, MLOps, distributed training, and edge deployment.

• GPU computing
• TPU
• AI accelerators
• model serving
• inference optimization
• MLOps
• +15

Teaching machines to see — recognition, detection, segmentation, diffusion-based generation, video understanding, and 3D vision.

• image recognition
• object detection
• image segmentation
• image generation
• diffusion models
• GANs
• +11

Aligning, governing, and red-teaming AI systems — bias, interpretability, regulation, RLHF, and the policy debates that shape deployment.

• AI alignment
• AI bias
• fairness
• interpretability
• explainability
• XAI
• +15

Working with text and speech — classification, translation, summarization, voice AI, conversational systems, and multilingual NLP.

• natural language processing
• text classification
• sentiment analysis
• named entity recognition
• machine translation
• summarization
• +11

AI that acts in the physical world — robot learning, RL for control, sim-to-real, manipulation and locomotion, humanoids, and world models.

• robotics
• robot learning
• reinforcement learning for robotics
• sim-to-real transfer
• manipulation
• locomotion
• +10