<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Practitioner Tips on Foresight Signals</title><link>https://foresightsignals.com/en/tips/radar/</link><description>Recent content in Practitioner Tips on Foresight Signals</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Thu, 12 Nov 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://foresightsignals.com/en/tips/radar/index.xml" rel="self" type="application/rss+xml"/><item><title>The Home Repo Pattern</title><link>https://foresightsignals.com/en/tips/radar/r-20261113-1/</link><pubDate>Thu, 12 Nov 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261113-1/</guid><description>Gather your skills and AGENTS.md in one repo and point every other repo to it — one session can then work with the whole system in view</description></item><item><title>You Can't Tell What's Real Anymore. Here's What Comes Next - Cristóbal Valenzuela, Runway</title><link>https://foresightsignals.com/en/tips/radar/r-20261103-1/</link><pubDate>Tue, 03 Nov 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261103-1/</guid><description>Treat video generation as next-frame prediction and judge it on physics and geometry, and you measure how well a model understands the world, not whether it suits someone&amp;rsquo;s creative taste</description></item><item><title>How To Use Decision Models for Evals</title><link>https://foresightsignals.com/en/tips/radar/r-20261023-1/</link><pubDate>Tue, 20 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261023-1/</guid><description>Don&amp;rsquo;t settle for a decision model&amp;rsquo;s default 0.50 threshold — tune it on your own labelled data and the same model catches failures it used to miss</description></item><item><title>How to Know When an AI Agent Can Work Unsupervised — Evan Kormos, Coinbase</title><link>https://foresightsignals.com/en/tips/radar/r-20261020-1/</link><pubDate>Mon, 19 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261020-1/</guid><description>Draw the process as a DAG and turn each node&amp;rsquo;s check into an evaluation — you can see which steps to hand to an agent and which steps humans still hold</description></item><item><title>Smarter Models Killed the Harness. Here's What's Left — Jediah Katz, SpaceXAI</title><link>https://foresightsignals.com/en/tips/radar/r-20261020-2/</link><pubDate>Mon, 19 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261020-2/</guid><description>Strip out the coding aids in your harness that the model has already learned, and invest in cloud runtime, permission scope, and cost per task — the score stays the same but the cost changes</description></item><item><title>The Model Swap Workshop: GPT-5.5 vs Kimi vs DeepSeek vs Mistral — Pamela Fox, Microsoft</title><link>https://foresightsignals.com/en/tips/radar/r-20261020-3/</link><pubDate>Mon, 19 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261020-3/</guid><description>Plug your own scenarios straight into several models and run them — you see the real differences in your app that benchmarks cannot tell you</description></item><item><title>I Spent Two Years Building the Median — Hariprasad Sudharshan</title><link>https://foresightsignals.com/en/tips/radar/r-20261016-2/</link><pubDate>Wed, 14 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261016-2/</guid><description>Before you build, name one thing that is new and yours and put it on the roadmap — and avoid spending two years on the same output as everyone else</description></item><item><title>Paving the Road to Production: Governed Enterprise AI Agents — Harish Gaggar, Credit Karma</title><link>https://foresightsignals.com/en/tips/radar/r-20261016-3/</link><pubDate>Wed, 14 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261016-3/</guid><description>Before you build agents, lay down registration, a gateway, memory, tracking and PII blocking as one set — so every agent passes through the same control path</description></item><item><title>Agents in Orbit: Workflows, Not Autopilot — Eduardo Simon &amp; Javier Casal, Flywire</title><link>https://foresightsignals.com/en/tips/radar/r-20261013-1/</link><pubDate>Mon, 12 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261013-1/</guid><description>Define agents as YAML workflows, not free-running loops — per-step permissions show up in code review and runs become predictable</description></item><item><title>Reliable, Smart Workflows for Finance, Wealth and Insurance — Karthic Rao, Perpetual</title><link>https://foresightsignals.com/en/tips/radar/r-20261013-2/</link><pubDate>Mon, 12 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261013-2/</guid><description>Leave control flow, evidence and outbound actions to deterministic code and confine the model to typed decision points — ambiguous requests no longer break the workflow, and compliance can sign off on it</description></item><item><title>AI News: Haiku-5.5, Visual ChatGPT, Grok's New Trick, and more!</title><link>https://foresightsignals.com/en/tips/radar/r-20261013-4/</link><pubDate>Fri, 09 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261013-4/</guid><description>Push repetitive work such as summarizing and classification down to Haiku 5.5 — at about 21 cents per task, it cuts cost to roughly a quarter of Sonnet&amp;rsquo;s</description></item><item><title>Why 80% Reliability Isn't Good Enough — Felipe Blanes, Amazon AGI Lab</title><link>https://foresightsignals.com/en/tips/radar/r-20261009-3/</link><pubDate>Fri, 09 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261009-3/</guid><description>Replace static evals with an eval flywheel driven by customer signals — turn post-launch hope into data</description></item><item><title>Brains vs Hands: How to Run AI Agents Safely in Production — Viren Baraiya</title><link>https://foresightsignals.com/en/tips/radar/r-20261009-5/</link><pubDate>Thu, 08 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261009-5/</guid><description>Let the LLM only decide what to do next and hand execution to a deterministic harness — cluster restarts then run the same way every time</description></item><item><title>How a Remote Company Builds AI Fluency — Em Shreve, Automattic</title><link>https://foresightsignals.com/en/tips/radar/r-20261009-4/</link><pubDate>Thu, 08 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261009-4/</guid><description>Don&amp;rsquo;t just roll out tools — clear two full weeks, half workshops and half real work, and even non-engineers come back with tools people use</description></item><item><title>Advanced Workshop: Mastering AI Observability — Doug Guthrie, Braintrust</title><link>https://foresightsignals.com/en/tips/radar/r-20261006-3/</link><pubDate>Tue, 06 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261006-3/</guid><description>Attach scores and Topics labels to your traces, and you can pull unknown failures out of accumulated logs and feed them back into evaluation datasets and PRs</description></item><item><title>How Many Credentials Should Your AI Agent Have? Zero. — Jim Clark, Docker</title><link>https://foresightsignals.com/en/tips/radar/r-20261006-4/</link><pubDate>Tue, 06 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261006-4/</guid><description>Strip every credential out of the agent sandbox and control tools through a single MCP gateway: if something goes wrong, the damage stays inside that one task</description></item><item><title>What Is an Inference Engine, Anyway? — Charles Frye, Modal</title><link>https://foresightsignals.com/en/tips/radar/r-20261006-5/</link><pubDate>Tue, 06 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261006-5/</guid><description>Break the inference engine down by workload, scheduler and cache, and you can pinpoint the causes of latency and throughput problems from the dashboard alone</description></item><item><title>The /implement-spec Skill</title><link>https://foresightsignals.com/en/tips/radar/r-20261006-2/</link><pubDate>Mon, 05 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261006-2/</guid><description>Implement a spec&amp;rsquo;s tickets in parallel with subagents, and hand off opening a session per ticket and tracking order to a single orchestrator</description></item><item><title>AI News: Dots, GPT-6.1 Sol, Sonnet 5.5, Gemini 4, and everything you need to know</title><link>https://foresightsignals.com/en/tips/radar/r-20261006-6/</link><pubDate>Fri, 02 Oct 2026 00:00:00 +0000</pubDate><guid>https://foresightsignals.com/en/tips/radar/r-20261006-6/</guid><description>Compare an always-on agent that picks its own tools without model selection against the new half-price models, and rework the model-and-cost mix in your workflow</description></item></channel></rss>