More from Eugene Yan
A sandboxed target, inputs that influence task difficulty, tools, and a grader.
Context as infra, taste as config, verification for autonomy, scale via delegation, closing the loop.
An eventful year of progress in health and career, while making time for travel and reflection.
Label some data, align LLM-evaluators, and run the eval harness with each change.
Based on what I've learned from role models and mentors in Amazon
More in AI
Many farmers and ranchers in eastern Colorado are curious about artificial intelligence, but they don’t have a local tech-savvy person to show them how to use it. Xue “Alice” Dong has made that her job. Dong, an IEEE senior member, is an extension regional specialist with the Colorado State University Extension, CSU’s outreach arm. She works in Logan County and serves a 10-county farming area in Colorado’s northeast corner. Her job is to turn university research into practical help for the people who live there. Xue “Alice” Dong Employer Colorado State University Extension in Logan County Title Extension regional specialist Member grade Senior member Alma maters Inner Mongolia Normal University, in Hohhot; Beijing Normal University; State University of New York College of Environmental Science and Forestry, in Syracuse; Colorado State University Global She built the AI-Powered Agritourism Chatbot (AgriTourismAI.com), which helps farmers and ranchers launch and run “agritourism” businesses. Agritourism is when farmers open their property to visitors for fun, education, or shopping events. Familiar examples include pick-your-own fruit orchards, lessons on how to grow vegetables, and sales of handmade products. Dong also teaches workshops where she shows farmers and ranchers how they can use AI tools to help grow and manage their business. She says she sees herself as a link between the university and the countryside. “We [extension specialists] act as a bridge,” she says. “We bridge our campus to the rural communities we serve.” Many farmers and ranchers are older, and they typically have no technical training. Often their children and grandchildren have left the farm to work in a city or otherwise aren’t around to teach them about new tools. Without help, the benefits of AI could pass the farmers by. Dong’s workshops give them a way in. Dong was named a finalist in the 2026 GenAI for Good Challenge for her AgriPivot AI Agent. The challenge was organized by IEEE Humanitarian Technologies and theU.N.’s International Telecommunication Union, and her project was selected from 318 teams across 79 countries. She also won the 2026 Award for Innovation from the National Extension Technology Community (NETC), and both the 2025 national and Western regional Educational Technology Awards from the National Association of Community Development Extension Professionals (NACDEP). She has secured seven competitive grants totaling more than US $250,000 as principal investigator or co-principal investigator, including funding supported by the Department of Agriculture’s National Institute of Food and Agriculture. An early love of design and problem-solving Dong was born in Beijing. Both her parents are doctors. As a child, she liked drawing new and inventive things. Sketching, she says, lets her picture solutions to problems in her head and put them on paper as practical designs. That habit grew into a way of relating to her surroundings. “I’ve always had a passion to create something tangible that can solve problems in the real world,” she says. She doesn’t credit a single mentor for the direction her life has taken. Instead, she says all her teachers helped lay the groundwork for her career. She attended the Inner Mongolia Normal University, in Hohhot, an institution that trains teachers. She earned two bachelor’s degrees there in 2013—one in geographical science and one in education. She then earned a master’s degree in curriculum and instruction from Beijing Normal University in 2016. That year, she moved to the United States to continue her graduate studies at the State University of New York College of Environmental Science and Forestry, in Syracuse. She earned a master’s degree in environmental biology in 2020. From outreach to extension work After earning her second master’s degree, she joined Paul Smith’s College, in Paul Smiths, N.Y., as an education and outreach coordinator. She managed several national watershed projects to protect rivers, lakes, and aquifers. Watersheds are areas of land from which water drains into bodies of water. She also undertook invasive-species-prevention projects, wrote grants to secure project funding, and designed two educational websites and several outreach publications. One website, Maintain the Chain, helps lake residents, boaters, and visitors from around the world keep invasive species out of the Fulton Chain of Lakes in New York’s Adirondack Park. In 2022, she joined Stewards of the Coast and Redwoods as program manager. The nonprofit, based in Guerneville, Calif., works with California State Parks to protect, restore, and educate the public about the parks along Sonoma County’s coast and the Russian River area. The region is known for its redwood forests, wineries, and resort towns. She left in 2023 to join the CSU Extension as a regional specialist, a title she still holds. Her employer is part of the Cooperative Extension System, a nationwide network run by land-grant universities, schools founded with federal grants to teach farming and other practical subjects. Extension faculty and staff share what researchers at their universities learn with the citizens of those states. The cooperative system operates on three levels. It is overseen federally by the Department of Agriculture’s National Institute of Food and Agriculture, managed statewide through land-grant universities, and delivered locally by county offices and Extension agents across more than 3,000 counties. The work goes well beyond farming, Dong says, also covering community and economic development, health initiatives, and the 4-H program, which teaches children hands-on skills in agriculture, science, leadership, and citizenship. Extension faculty members’ performance is judged on their research, teaching, and service to the community, Dong says—similar to the way professors in academia are evaluated. The difference, she says, is the audience: Professors teach students; most extension specialists teach adults, including farmers, ranchers, and community leaders. The setting is different, too. Dong doesn’t commute from the campus to the countryside; she lives and works in it. Her office is in Sterling, a small rural town. But her projects serve the whole state, and many of the people she helps live one to four hours’ drive away. So, most of her work with farmers and ranchers happens by email, phone, and virtual meetings. County fairs, farmers markets, and agricultural events give her chances to meet large groups of them in person. She can’t just hand Colorado residents research findings and expect them to figure out the information they need, she says. “We need to translate that research into something more practical, and easier to use, that farmers and ranchers can understand immediately,” she says. Dong also focuses not on only on whether people use AI, but also on how they use it. In 2026, she and Paul Hill from Utah State University Extension published “From Access to Governance: Generative AI Literacy for Extension Professionals and Small Businesses, and a Governance-Aware Design Framework” in the journal AI. The paper offers a framework for helping Extension professionals and small businesses adopt generative AI responsibly. How her AI chatbot helps farmers and ranchers In 2025, Dong earned a third master’s degree, in artificial intelligence and machine learning, from CSU Global. She immediately put those skills to work on agritourism, she says. She built the chatbot in Python, a popular programming language. It answers typed questions, and its user interface looks much like ChatGPT’s, but it works differently. Under the hood, it uses retrieval-augmented generation, a technique that has the AI look up information in a set of trusted documents before it answers. Dong curated that collection from Extension publications, so the chatbot’s answers rest on authoritative sources and are less likely to be made up. She presented the project at NeurIPS 2025 (Conference on Neural Information Processing Systems). The code is openly available on GitHub, so other Extension offices can adapt it. The bot runs in a web browser rather than as a phone app. Many farmers she works with don’t own a smartphone, Dong says, but nearly all have a computer, tablet, or other gadget that can connect to a website. The bot works on any of them, with no app to download. Extension specialists “act as a bridge. We bridge our campus to the rural communities we serve.” Farmers often ask the bot how to attract visitors and how to write a business plan. A farmer thinking about opening a corn maze, for example, can ask how to price admission, what kind of insurance to carry, and which permits to get. The bot also gives advice on managing risk. One of her test sites was Prairie Ridge Buffalo Ranch, in eastern Colorado. Dong documented how the ranch put AI to work in a peer-reviewed case study, “Revolutionizing Agritourism with AI and Machine Learning: A Case Study of Prairie Ridge Buffalo Ranch in Colorado,” published in 2025 in CABI Tourism Cases. Surveys and interviews she conducted with other users showed that people find it’s easy to use the bot, learn from it, and act on its advice, she says. She is preparing the full results for publication. Her work extends beyond agritourism. In 2026, she contributed an AI assistant to Project Pythia, an open-science program funded by the U.S. National Science Foundation and led by the National Center for Atmospheric Research (NCAR). The AI assistant turns raw airport weather report data into plain-language guidance for farmers and other non-technical users, and it tells users when data is missing rather than guessing. She also teaches workshops on AI tools such as ChatGPT, Claude, and Gemini. She has taught her workshop 20 times to more than 200 entrepreneurs and business owners, who came from 32 Colorado counties. In follow-up surveys, 90 percent of participants reported trying at least one new AI task in their business. Some of the workshops take place at a community center run by the Logan County Economic Development Corp., in Sterling. She has also co-hosted the workshops with the Rural Community Assistance Corp., Colorado Startup Week, and Boulder Startup Week. And she occasionally presents them as webinars, so people don’t have to travel. Her outreach has taught her that many farmers are eager to try new tools, she says. She shows them how simple her chatbot is to use. She also offers advice to high school students drawn to the field. Extension specialists each have a specialty, such as livestock, natural resources, food systems, or 4-H. Students should pick one based on their passion, she says. Land-grant universities offer related programs at the bachelor’s and master’s levels. For graduate students, she adds, Extension offers a third career path beyond academia and industry: “You apply advanced skills directly to the communities that need them.” How IEEE membership supports her career Dong joined IEEE in 2025 and soon was elevated to senior member. Her professional home is the IEEE Society on Social Implications of Technology. She also serves as vice treasurer of the IEEE Denver Section’s Young Professionals group. IEEE supports all three parts of her job, she says. For research, she presented “Entrepreneur AI Agent: Empowering Small Businesses and Bridging the Digital Divide through Context-Aware Generative AI” at the 2026 IEEE Global Humanitarian Technology Conference (GHTC). For teaching, she has offered to bring her AI workshop to Denver Section members and local entrepreneurs. For service, she reviews papers for IEEE conferences. IEEE conferences matter most to her, she says: “Through them, I can present my research and network with other researchers in the field.” She attended the IEEE World Forum on Public Safety Technology in Boulder, Colo., in September 2026. The event, she says, opened her eyes to many new projects underway across IEEE. Still, she says, her work always points back to the farms and ranches of eastern Colorado. Its people are curious, she says, but they can’t make the leap to new technology alone. “They really need this kind of a bridge,” she says.
Early in the coronavirus pandemic, as I sat at home channeling my anxiety into publishing daily essays on this newsletter (a streak I kept up ... Read more The post In Search of the Deep Life appeared first on Cal Newport.
Prefix caching helps coding agents stay fast as conversations grow. Prompts, tool calls, and inference routing determine how much work the model can reuse.
The tool we created to build understanding and mastery for our own internal work at Answer.AI
Internalize this graph, we post it on this blog a lot. If trends continue, China will have 10x the electricity of the US in 10 years. Whenever you think about geopolitical strategy, always keep it in mind, because whatever your plans are must consider this. Here’s a strategy that would make sense for America long term. Sell as many AI chips into China as possible, use the money to purchase solar panel and battery factories, and build out electricity capacity as quickly as possible. Fundamentally, compute is a function of electricity. That’s the sort of long term thinking that gives America a chance. With robotic labor, immigrants like Jensen and Elon, a business friendly culture, and an incredible chunk of land, it’s not so clear who is ahead in 50 years. Regardless, I do know this strategy would lead to massive prosperity for the citizens of both countries. However, that’s not close to what we see happening. We see a lot of short term zero sum thinking in action. It’s very scary what the US “AI race” mentality leads to. When exactly does the race end? Whatever capabilities you have today will eventually be matched, unless… …you believe that you can use ASI in a decapitating first strike. I’m worried that this is what these people see as the only way out. It probably doesn’t work, can you outthink steel and drone production? I don’t see how, see my critiques here. But I’m not totally sure, I’m not a superintelligence. And if it does work, that may be the end. My likelihood of imminent WWIII went up a lot with the nationalization of NVIDIA.