Nick Lloyd

The harness is no longer a wrapper around the model. The harness is part of the model's effective parameters. The post training process embeds the harness's tool surface, schema shapes, memory rituals, citation contracts, and system prompt structure into the model's instinct set. You can take the weights to a different harness, but you cannot take the instincts. The instincts only fire when the harness presents the world the way the post training presented it. Also, the matched pair is not static. The right harness for a model in March is not the right harness for that model's successor in October! Once again, if you want to stay at the edge, you have to delete most of your code when a new model is released... LLMs eat scaffolding for breakfast!

Every abstraction shift in software history made devs more productive by raising the level of intent. This is the next step: from writing code to orchestrating systems that write code (building "the factory" for your code). The unsolved problem isn't generation but verification. That's where engineering judgment becomes your highest-leverage skill. To truly scale, think "factory model" - orchestrate fleets of agents like a production line: clear specs as blueprints, TDD for quality control, strong architecture to amplify leverage.

Some people today are discouraging others from learning programming on the grounds AI will automate it. This advice will be seen as some of the worst career advice ever given. I disagree with the Turing Award and Nobel prize winner who wrote, “It is far more likely that the programming occupation will become extinct [...] than that it will become all-powerful. More and more, computers will program themselves.”​ Statements discouraging people from learning to code are harmful! In the 1960s, when programming moved from punchcards (where a programmer had to laboriously make holes in physical cards to write code character by character) to keyboards with terminals, programming became easier. And that made it a better time than before to begin programming. Yet it was in this era that Nobel laureate Herb Simon wrote the words quoted in the first paragraph. Today’s arguments not to learn to code continue to echo his comment. As coding becomes easier, more people should code, not fewer! Over the past few decades, as programming has moved from assembly language to higher-level languages like C, from desktop to cloud, from raw text editors to IDEs to AI assisted coding where sometimes one barely even looks at the generated code (which some coders recently started to call vibe coding), it is getting easier with each step. I wrote previously that I see tech-savvy people coordinating AI tools to move toward being 10x professionals — individuals who have 10 times the impact of the average person in their field. I am increasingly convinced that the best way for many people to accomplish this is not to be just consumers of AI applications, but to learn enough coding to use AI-assisted coding tools effectively. One question I’m asked most often is what someone should do who is worried about job displacement by AI. My answer is: Learn about AI and take control of it, because one of the most important skills in the future will be the ability to tell a computer exactly what you want, so it can do that for you. Coding (or getting AI to code for you) is a great way to do that. When I was working on the course Generative AI for Everyone and needed to generate AI artwork for the background images, I worked with a collaborator who had studied art history and knew the language of art. He prompted Midjourney with terminology based on the historical style, palette, artist inspiration and so on — using the language of art — to get the result he wanted. I didn’t know this language, and my paltry attempts at prompting could not deliver as effective a result. Similarly, scientists, analysts, marketers, recruiters, and people of a wide range of professions who understand the language of software through their knowledge of coding can tell an LLM or an AI-enabled IDE what they want much more precisely, and get much better results. As these tools are continuing to make coding easier, this is the best time yet to learn to code, to learn the language of software, and learn to make computers do exactly what you want them to do. [Original text: deeplearning.ai/the-batch/issu… ]

2025 will be the year of agentic systems The pieces are falling into place: computer use, MCP, improved tool use. It's time to start thinking about building these systems. At Anthropic, we're seeing a few best practices emerge - we wrote a blog post with our findings:

Coding with AI is akin to the current version of FSD - you can get pretty far, but keep your hands on the wheel. Won't be that way for long - for either...

Adding evals and monitoring to an agentic app has been an 'amazing grace' moment. I was blind, but now I see.

It is time to deprecate HumanEval! 🧑🏻‍💻 @BigCodeProject just released BigCodeBench, a new benchmark to evaluate LLMs on challenging and complex coding tasks focused on realistic, function-level tasks that require the use of diverse libraries and complex reasoning! 👀 🧩 Contains 1,140 tasks with 5.6 test cases each, covering 139 libraries in Python. 📊 Uses Pass@1 with greedy decoding and Elo rating for comprehensive evaluation. 🏆 Best model is GPT-4o 61.1%, followed by DeepSeek-Coder-V2. 🥈 Best open Model is DeepSeek-Coder-V2 with 59.7%, better than Claude 3 Opus or Gemini. 👥 Tasks are created in a three-stage process, including synthetic data generation and cross-validation by humans. 🧱 Evaluation framework and Docker images available for easy reproduction 🔜 Plans to extend to multilingualism. Blog: hf.co/blog/leaderboa… Leaderboard: huggingface.co/spaces/bigcode… Code: github.com/bigcode-projec…

ok I'm used to Belgium 🇧🇪 not being on the list, but not Canada 🇨🇦 ?? 😅 #available_regions" target="_blank" rel="nofollow noopener">ai.google.dev/available_regi…

After years of avoiding, I finally decided to get a @Medium membership. In just a month, it has exponentially expanded my learning opportunities. The internet feels like what it was supposed to be: a hub for quality knowledge and perspectives from domain experts.

Dang it - was planning on working tonight, but @xlntsound dropped their new Quest for Bass. Best laid plans...

How are companies *𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲* using AI? We surveyed 1500+ devs, builders, and business leaders to explore production use cases, the models, infrastructure, and tools people are using, and much more. Dig into the new report: retool.com/reports/state-…

You’ll often hear people say someone has good “work ethic” if they’re putting in long hours. But 60, 70, 80 hours a week doesn’t equal work ethic. 60, 70, 80+ hours a week simply equals 60, 70, 80+ hours a week. Work ethic is about showing up, being on time, being reliable, doing what you say you’re going to do, being trustworthy, putting in a fair day’s work, respecting the work, respecting the customer, respecting the organization, respecting co-workers, not wasting time, not making work hard for other people, not creating unnecessary work for other people, not being a bottleneck, not faking work. Work ethic is about being a fundamentally good person that others can count on and enjoy working with. Works a lot ≠ has good work ethic.

@johnlepp @LMStudioAI @continuedev ¯\_(ツ)_/¯

@nickleplated @LMStudioAI @continuedev For as long as I’ve known you nick - I feel like you’ve always needed a beefier computer…

Had a lot of fun playing with @LMStudioAI + @continuedev tonight... Feels like imagination within reach! Though I _may_ need a beefier machine soon...

Arguing with an engineer is a lot like wrestling in the mud with a pig, after a couple of hours you realize the pig likes it.

Welcome to the Workable integration ecosystem, @Apideck 🤝 Apideck provides the infrastructure building blocks for B2B SaaS companies to enable them to build and market integrations through one platform.