Ben Doessel
5.8K posts
You thought the fun was over? 🏈 This weekend, video takes center stage on the timeline. We’re awarding $1M, $500K, and $250K to the top three videos about @grok, created with Imagine 1.0.
1/5 I'm a cardiologist. I have spent twenty years watching cholesterol destroy arteries, trigger heart attacks, and kill people I care about. Today, Eli Lilly presented data that may begin to end that era. VERVE-102. A single infusion. One dose. It uses base editing to permanently turn off the PCSK9 gene in your liver. Presented today at the European Atherosclerosis Society Congress: 88% reduction in PCSK9. 62% reduction in LDL cholesterol. Sustained up to 18 months. No treatment-related serious adverse events. One infusion. Not daily pills you forget to take. Not monthly injections. One dose — and your cholesterol may stay low for the rest of your life.
🇺🇸 A robot just worked an 8-hour shift. Fully autonomous. Sorted packages at human speed and then diagnosed its own maintenance issues Figure AI's Helix-02 replaced 109,000 lines of code with a neural network trained on human movement. One system. Legs, arms, fingers, eyes. All running together like a person. Get this: in a warehouse in America, a robot just clocked out after a full shift. And immediately clocked back in because it doesn't sleep.
Gradient descent for SKILL.md files sounds interesting, maybe a bit complex but it's becoming a real part of agent harness. SkillOpt is one of the first papers to treat markdown skill files as trainable parameters and provides a proper optimization framework for them. A few things I learned that you should consider too. 1. The validation gate is the only thing that matters in a self-editing loop. Held-out set, strict improvement, ties rejected. End-to-end, their best skills land with 1 to 4 accepted edits total. If your "self-improving agent" is accepting most of what it proposes, you're shipping slop. 2. Bounded edits are better than full rewrites. 4 to 8 edits per step is the sweet spot. Remove the budget and performance collapses. This is the textual analog of learning rate, and it transfers to any LLM-as-author loop. If you're using an agent to refactor your docs, your prompts, or your skills, cap the diff size. 3. Compactness wins. Median final skill: ~920 tokens. Skills do not need to be long. They need to be high-signal. Most skill files I see are bloated because length feels like effort. It isn't. 4. The harness is becoming less important; the skill is becoming more important. A Codex-trained skill ported into Claude Code hit +59.7 points on SpreadsheetBench. Procedural knowledge is more general than the runtime that produced it. 5. Frozen model + trained context is the practical adaptation. GPT-5.4-nano with a SkillOpt'd skill ≈ frontier behavior on procedural benchmarks. Cheaper, portable, inspectable, zero inference-time cost. This is the answer to "how do we adapt a frontier model for our domain" for almost everyone who isn't training their own models. 6. Verification is the bottleneck. Every gate in this paper depends on an auto-grader. That works for benchmarks. It fails for writing, design, and strategy, exactly the open-ended work we want to automate. Whoever builds the verifier for open-ended tasks owns the next stage. There are also two leassons I learned while shipping v2.3.0 of my Context Engineering Agent Skills repo, measured across composer-2, claude-opus-4-7, gpt-5.5, and gemini-3.1-pro via the @cursor_ai SDK: - Description and body are two different surfaces. The router only sees the description. The agent sees the body once activated. They can quietly disagree, and only end-to-end task tests catch it. - Aggregate accuracy is the wrong unit. When I rewrote three descriptions, the corpus average moved ~1pp. Individual skills moved 23–25pp. Per-skill effect size is where the action is. Also, in Feb 2026 I shared a piece called Personal Brain OS arguing that the markdown file is a first-class substrate for agent state. SkillOpt is the optimizer-shaped version of that same argument: not "store memory in files" but "treat files as trainable parameters with proper optimization machinery around them." That's the move from static to measured. The fast/slow split they describe already lives implicitly in the digital-brain-skill repo: - voice-guide and tone-of-voice.md are slow-state (rarely touched) - posts.jsonl and bookmarks.jsonl are fast-state What SkillOpt adds that I didn't have is a protected section invariant, a structural guarantee that fast edits cannot overwrite slow lessons. Removing that mechanism cost them 22 points on SpreadsheetBench. Worth borrowing. If you're building agents, SkillOpt: Executive Strategy for Self-Evolving Agent Skills is a good paper to read: arxiv.org/pdf/2605.23904
Most AI-generated brand videos still feel generic. So I tested if Claude Code/Codex + Higgsfield AI could create a brand-grounded content factory. Turns out, yes. Here's the workflow I used to create this video for @touchland : 1. Research the brand, products, ICP, positioning, and visual style 2. Turn it into a brand bible 3. Generate video concepts for different customer segments 4. Pull real product images from the website 5. Preprocess them with Nano Banana 2 6. Write a storyboard with consistent scenes/props 7 .Send assets + storyboard to Higgsfield 8. Have Claude self-QC frames and iterate The bigger opportunity: Brands can run this across hundreds/thousands of SKUs and generate product videos grounded in real product info, brand assets, and customer segments. I turned this into a reusable Skill. Let me know if you want access.
