Newsletter archive
Applied Signals Weekly: 2026-06-21
Jun 15, 2026 to Jun 21, 2026
In this week’s edition
This week’s edition surfaces where agentic systems face the sharpest pressure points: reliability under coordination, evaluation discipline, and the practical mechanics that keep multi-agent behavior stable over time.
- A near-autonomous AI chemist improves a challenging reaction in medicinal… — OpenAI and Molecule.one show how a near-autonomous AI chemist using GPT-5.4 improved a key drug-making reaction, advancing medicinal chemistry research.
- MemoryWAM: Efficient World Action Modeling with Persistent Memory — Robust robotic manipulation in the real world requires not only an understanding of the current observation, but also memory and dynamics modeling. World action models (WAMs) possess these capabilities by jointly…
- Two labs started dreaming, and they built two different architectures — Anthropic and OpenAI both shipped 'dreaming' for AI memory in May and June 2026, and they built opposite architectures. A look at what each lab shipped, what the empirical literature says, and what to do if you are…
- What is an agent harness? Why harnesses are replacing agent frameworks — Agent harnesses are replacing frameworks as the real product surface for reliable AI agents, shifting the work from prompt tuning to loops, tools, traces, evals, and operational metrics.
- Current World Models Lack a Persistent State Core — World models are increasingly regarded as a decisive step toward artificial general intelligence, yet modeling the physical world demands more than rendering convincing frames on demand: it requires an internal world…
- Is it agentic enough? Benchmarking open models on your own tooling
- Meet PXI: the AI engineering agent inside Phoenix — An AI engineering agent built into Phoenix. It works like a coding agent, just point it at your telemetry instead of a source tree.
Worth watching
- evaluation + tool-use: hype or repeatable pattern?
- planning + tool-use: hype or repeatable pattern?
- evaluation + memory: hype or repeatable pattern?
Read the full analysis
- Grite, CEO-Bench, and ADE quantify coordination and stability gaps (weekly trend)
- Jun 21, 2026 MemoryWAM hybrid memory and MosaicLeaks query-log leakage results
- Jun 20, 2026 LedgerAgent and OpenAI
- Jun 19, 2026 AgenticRei runtime governance, grite event logs, SIGMA compositional agents
- Jun 18, 2026 NAVI-Orbital, OmniAgent, and CEO-Bench agent robustness results
- Jun 17, 2026 MemSlides, IC2PP equilibria, and MemTrace evidence-use evaluation
- Jun 16, 2026 PhoneHarness, DR-DCI, DeepRoot and budgeted web-agent evaluation
- Jun 15, 2026 WebDecept safety evals, Orchestra-o1 OmniGAIA, TwinBI state coherence
Subscribe
Get Applied Signals weekly newsletter in your inbox