Newsletter archive
Past issues of the Applied Signals weekly email, newest first. Subscribe to get the next one.
- Week ending Nov 15, 2026
How Amplitude ships product improvements in 1–2 days with OpenAI · Architecture patterns for fast session startup on Amazon Bedrock AgentCore… · BAIR at CoRL 2026
- Week ending Nov 01, 2026
Browser Use in Managed Deep Agents · Managed Deep Agents v0.10: Browser Use and Async Subagents · rTaste: Do Research Agents Have Taste? Evaluating Early Stage Research Judgment
- Week ending Oct 18, 2026
Grounding AI agents in biological reality: Building blocks for autonomous wet… · Introducing Reflection Beam: an open-weight model for agentic coding on AWS · October 2026 LangChain Newsletter: Everything we shipped at Interrupt NYC &…
- Week ending Oct 11, 2026
Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training… · Dex-One2Many: Learning Dexterous Manipulation from a Single Human Demonstration · ICYMI: What landed for AI builders in September 2026
- Week ending Oct 04, 2026
A model guide for the GPT-6 family · ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization · ROWBench: Do Video Models Render What the Program Specifies?
- Week ending Sep 27, 2026
Coding Agents for Generalized Task and Motion Planning Problems · Rolling-WAM: World Action Models with Rolling Imagination · Advancing Private AI Compute with secure, server-side memory
- Week ending Sep 20, 2026
An Empirical Study of Harness Design for Coding Agents · Don't Mask the Environment: Observation Supervision Changes How Agents Explore… · How Fyxer built an AI executive assistant people trust
- Week ending Sep 13, 2026
Biology-in-the-loop: Amortized Adaptive Hit Discovery in CRISPR Screens · Code mode: Why your agent should code · Cognition helps Devin test its own work with GPT‑6 Astra
- Week ending Sep 06, 2026
Agent cost management is about more than the model · Terminal-Universe: Turning Agent Trajectories into Scalable Terminal… · Give Your Coding Agents a Memory You Own
- Week ending Aug 30, 2026
FlashVLA: Streaming Action Decoding for Fast and Asynchronous VLA Inference · How Signal found two hidden retry loops in our production agent Alyx · UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City
- Week ending Aug 23, 2026
AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive… · DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation · MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill…
- Week ending Aug 16, 2026
AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design · Behavioral Reprogramming of Open-Weights Models: Cognitive Plasticity and… · AI agent guardrails vs. evals: How to build more reliable agent systems
- Week ending Aug 09, 2026
DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for… · How cheap models changed multi-agent economics · The Low Frequency Trap: Video Language Models Fail at Simple Event Bookkeeping
- Week ending Aug 02, 2026
ACE-Data-0: Human-Centric Ambient Capture as Embodied Data Engine · AskChem: Claim-Centered Infrastructure for Chemistry Literature Synthesis · From Signal to PR: What if your agents got better every time they failed?
- Week ending Jul 19, 2026
Hierarchical Denoising For Multi-Step Visual Reasoning · Inside Cursor’s agent factory: how it verifies AI-written code · Kiro CLI observability: trace and evaluate agent changes with Arize Skills
- Week ending Jul 12, 2026
ChatGPT is now a partner for your most ambitious work · Own the loop: A field guide to agent harnesses · The agent is the user now: lessons from the founder of WorkOS
- Week ending Jul 05, 2026
Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on… · ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration · SkillOpt: Agent skills as trainable parameters
- Week ending Jun 28, 2026
Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and… · Build real agentic apps using CUGA: two dozen working examples on a lightweight… · Codex-maxxing for long-running work
- Week ending Jun 21, 2026
A near-autonomous AI chemist improves a challenging reaction in medicinal… · MemoryWAM: Efficient World Action Modeling with Persistent Memory · Two labs started dreaming, and they built two different architectures
- Week ending Jun 14, 2026
Bring production agent traces from Arize into Databricks Unity Catalog · PostgresFS vs. SQL skills: should AI agents fake a filesystem? · RoboProcessBench: Benchmarking Process-Aware Understanding in Vision-Language…
- Week ending Jun 07, 2026
HANDOFF: Humanoid Agentic Task-Space Whole-Body Control via Distilled… · TempoVLA: Learning Speed-Controllable Vision-Language-Action Policies · Building the AI factory for self-improving agents: What’s new in Arize AX
- Week ending May 31, 2026
How a Mid-Tier Enterprise SaaS Provider Automates Cloud Support Triage · ITBench-AA: Frontier Models Score Below 50% on the First Benchmark for Agentic… · SchGen: PCB Schematic Generation with Semantic-Grounded Code Representations
- Week ending May 24, 2026
Building a self-improving agent on a context graph of human disagreement · MagenticLite, MagenticBrain, Fara1.5: An agentic experience optimized for small… · How to build LLM-as-a-Judge evaluators that hold up in production
- Week ending May 17, 2026
Databricks brings GPT-5.5 to enterprise agent workflows · EntityBench: Towards Entity-Consistent Long-Range Multi-Shot Video Generation · From observability to context: What’s next for Arize Phoenix
- Week ending May 10, 2026
"OncoAgent: A Dual-Tier Multi-Agent Framework for Privacy-Preserving Oncology… · AI agent evaluation: How to test, debug, and improve agents in production · Agent harnesses have an expiration date
Subscribe
Get Applied Signals weekly newsletter in your inbox