ashb
ashb — researcher or builder tracked in the Angestrom contributor network.
Repositories · 6
apache/airflow
Apache Airflow - A platform to programmatically author, schedule, and monitor workflows
YoavMayer/babysitter-observer-dashboard
Real-time observability dashboard for AI agent runs: live task progress, journal events, and orchestration state. Published on npm.
arashbehmand/mom-llm
MoM Service: OpenAI-compatible API that orchestrates multiple LLMs in parallel and synthesizes their responses into superior answers. Get GPT-5, Claude, and Gemini working together. Features intelligent caching, multimodal vision support, cost tracking, and comprehensive observability.
periavi075/llm-tco-dashboard
dk-raas/dkai/agents/slashbay
Slashbay: issue webhook intake, cheap-LLM triage, Coder workspace berth, and coding-agent dispatch for the DataKnifeAI fleet.
niklasfrick/spark-dashboard
Real-time hardware and LLM inference monitoring — GPU, CPU, memory, and vLLM metrics streamed to a dashboard.
News · 13
Monitor and debug generative AI inference with SageMaker detailed metrics and Insights dashboard on CloudWatch
Amazon SageMaker AI provides fully managed real-time inference hosting for machine learning models. You deploy a model to a SageMaker endpoint backed by one or more compute instances, and SageMaker handles provisioning and scaling. SageMaker supports multiple endpoint architectures. This post focuses on the two most relevant to generative AI workloads with detailed observability: Single-model endpoints (SME) and Inference component (IC) endpoints.
The automotive software vulnerabilities hiding in your dashboard - Help Net Security
<a href="https://news.google.com/rss/articles/CBMilwFBVV95cUxQYWpzT0RKWDZaeG9iX3BSYmdxY1pwamZxdjZqdTZTeE56N3VyalJlMVZWbTZyZGRUSndIUWdoek1wRkRfSDhzOG5hakttY2tNVkhFSEhtNWtWTFA0cHcxVVBTY1UzOHFpdVd2SHRGREFkVGlyUWZGTGZYLWRWMU5XZmR5OFVHS25zR2pIaVlaM2dET19OdXNV?oc=5" target="_blank">The automotive software vulnerabilities hiding in your dashboard</a> <font color="#6f6f6f">Help Net Security</font>
Open-source Python library + no-code web dashboard for evaluating oncology AI models at clinical decision thresholds. [P]
<!-- SC_OFF --><div class="md"><p>Most classification metrics for oncology AI models (AUC, ICC, MAE) measure global agreement. They don't answer the question that actually matters at the point of care: how reliable is this model at the exact cutoff that decides whether a patient gets flagged, biopsied, or treated?</p> <p>I built <code>oncothresh</code> to evaluate models at a specific clinical threshold rather than in aggregate: sensitivity/specificity/PPV/NPV at the cutoff, bootstrap confidence
Cost attribution keeps finding waste that monitoring dashboards miss - how do others handle the gap?
<!-- SC_OFF --><div class="md"><p>I work on cost attribution for a large internal cloud fleet, and a pattern keeps repeating: our observability stack says everything is healthy, and the invoice says otherwise.</p> <p>Recent example: a batch worker that was fully green in monitoring — no errors, no alerts, normal resource graphs - but was costing ~$4,200/month sitting mostly idle because it was provisioned for a peak workload that moved to a different pipeline two quarters ago. Nobody's dashboard
AWS adds water withdrawal reporting to sustainability dashboard as data center water use comes under scrutiny - waterworld.com
<a href="https://news.google.com/rss/articles/CBMi9AFBVV95cUxPQjF6MWw1WUp4VURaQzYya05tczdmaUNiMjhUczhSZFAwdzFRZl9jdmRNYWRNMXhscUtmdm9vV3g0M1J0M2J6WmpTTjRfd0dNUjhDa0pseVBESDF3aWd0cHBYODBoRnN5MUJHM0w3bWdfTWpaMVFybEEyVDdhdnVKZkJpOE02RFYwZElGYWR5YjJZNHpaRjVVLXh0elRubnhLeDl5NmV0bTlQR3ZtT0d1QXZuSFQ1SzdSZGN3NVN0N2lsVzBqSC1aVUw3WkJtNnZ6N3Bmc2k4OGRvVklhRW1nSUNkdFBSdWo0UGR4MElVZ2N4VzBy?oc=5" target="_blank">AWS adds water withdrawal reporting to sustainability dashboard as data center water use comes unde
Weighing smoke: why AI visibility dashboards are mostly useless
<p>Article URL: <a href="https://betterthangood.xyz/blog/weighing-smoke/">https://betterthangood.xyz/blog/weighing-smoke/</a></p> <p>Comments URL: <a href="https://news.ycombinator.com/item?id=48819958">https://news.ycombinator.com/item?id=48819958</a></p> <p>Points: 11</p> <p># Comments: 0</p>
Introducing Mobile Layout for Amazon Quick dashboards
Teams that rely on dashboards for daily decisions often must pinch and zoom to interact with controls originally designed for larger displays. Checking revenue during a morning standup, reviewing pipeline metrics between meetings, or monitoring operations while traveling all require extra effort when the dashboard was built for a desktop screen. Mobile Layout for Amazon […]
Grafana Review 2026: Dashboard, Pricing, Login, Download, Free Plan & FAQs - Nubia Magazine!
<a href="https://news.google.com/rss/articles/CBMijgFBVV95cUxPQW4yR0U1cUJSV3FhUnpfMmNUTzFOREQ4dHNaelJqX1ZxUVJWVm1kM3NCSmlRQlhOX0lJZEtWTXZwekFWMVV2SjdUM2RzaXpLeW1sbEpZcTAyZ09zYXJqRlNXQVJqdzNhZlcyRVpqdDJvN0ozNWpXdVIxUm5YS2RaamlTczZlWm1iUUNIVXpR?oc=5" target="_blank">Grafana Review 2026: Dashboard, Pricing, Login, Download, Free Plan & FAQs</a> <font color="#6f6f6f">Nubia Magazine!</font>
Papers · 2
LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via a Proprioceptive Dashboard
Long-horizon tool agents are bottlenecked by how their context grows toward the limits of the context window. Recent systems make context management agent- or system-controlled, but they either learn a compression policy that discards evidence or manage context in a layer the agent never sees. We argue both leave a more basic gap unaddressed. Frontier language models are proprioceptively blind to their own context. From the prompt alone they cannot see how large, how old, or how used each block is, the signals a keep-or-drop decision needs. We hypothesize that competent context management is a
ImputeViz: A Visual Analytics Dashboard for Diagnosing Missing Data and Comparing Imputation Methods
Missing data is a persistent obstacle in scientific, social science, and public health research, often biasing analyses and placing accountability on analysts for how they handle missing values. We introduce ImputeViz, an integrated visual analytics dashboard that supports diagnosing missingness, configuring imputation models, and evaluating results. The system brings together widely used methods, including MICE, Random Forest, XGBoost, and kNN, within an interactive environment that makes missingness patterns explicit. To support geospatial reasoning, we introduce gKNN, a geographically infor
