AI news for platform & agent engineers
Ranked brief · refreshes every 2 hours
Agentic LLM systems have moved from linear prompt chains to graph-based workflows and, more recently, code-executing harnesses. Chains are simple but fragile, graphs add branching, retries, and validation, and code ha... Context & related coverage →
On-policy distillation aligns teacher feedback with a student's current behavior, yet acquiring useful supervision for multi-turn agents requires identifying both where guidance is needed and whether the student can s... Context & related coverage →
Agent harnesses repeatedly invoke reasoning after tool feedback, even when an earlier call has already fixed the next action. We introduce HarnessMM (Harness Minus Minus), a method for improving harness token efficien... Context & related coverage →
Agents and verifiers often rely on the text a tool prints, and that text can misrepresent what the tool actually did, even when the tool's computation is correct. We saw this happen in a production RAG research system... Context & related coverage →
Fine-tuning on datasets of unknown origin can create a backdoor in the trained model when the data includes poisoned examples. Difficult to localise and remove, backdoors can persist through safety training. While par... Context & related coverage →
Building Task-Oriented Dialogue (TOD) systems for high-value scenarios such as financial marketing requires large volumes of domain-specific dialogue data, which privacy regulations, annotation costs, and the need for... Context & related coverage →
With P/D disaggregation and Wide-EP on a GB300 NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 954K tokens/s per GPU at a P90 of 196 tokens/s per user, 7.9x our aggregated single-node result. Context & related coverage →
Agents waste time and context on clumsy searches, oversized results and stale data. Learn the retrieval techniques LangChain's Deep Agents harness uses to get the right context in. Context & related coverage →
Trainium Best for Inference? But now good for Training too, Native Trainium PyTorch, High Quality CI, TorchTitan, GPU Impossible Profiler Context & related coverage →