Eugene Yanayt
PulseAugur coverage of Eugene Yanayt — every cluster mentioning Eugene Yanayt across labs, papers, and developer communities, ranked by signal.
-
Eugene Yan shares insights on LLM system building and AI engineering trends
Eugene Yan presented key learnings from building with Large Language Models (LLMs) at the AI Engineer World's Fair 2024. The keynote, co-authored with others, focused on practical aspects of LLM system development, incl…
-
Eugene Yan explores challenges in evaluating abstractive summaries and detecting hallucinations
Evaluating abstractive summarization, which involves rephrasing source material rather than copying sentences, presents challenges, particularly in assessing relevance and factual consistency. While fluency and coherenc…
-
Eugene Yan builds Obsidian-Copilot to assist writing and reflection
Eugene Yan has developed a prototype tool called Obsidian-Copilot, designed to assist with writing and personal reflection. The tool functions by first chunking documents, prioritizing top-level bullets for notes, and t…
-
Eugene Yan compiles list of open-source LLMs for commercial use
Eugene Yan has compiled a list of open-source large language models (LLMs) that are available for commercial use. This resource was created to address the need for LLMs with commercial licenses, particularly for applica…
-
Eugene Yan explores LLM interfaces beyond chat for better user experience
Eugene Yan proposes alternative user experiences for interacting with large language models, moving beyond traditional chat interfaces. He suggests that for tasks like online shopping, users might prefer visual and inte…
-
Eugene Yan builds Raspberry-LLM to add AI smarts to low-resource Pico
Eugene Yan developed Raspberry-LLM, a project that integrates a large language model with a Raspberry Pi Pico, a low-resource microcontroller. This setup allows the device to interact with external data sources like RSS…
-
LLM-powered Biographies
Eugene Yan experimented with several large language models, including GPT-4, Claude-v1.2, and Cohere-xlarge, by asking them to generate his biography. He observed that while the models captured the general essence of hi…
-
Eugene Yan details how to write effective data labeling guidelines
Writing effective data labeling guidelines requires careful consideration of several key questions to ensure accuracy and consistency. These guidelines should clearly articulate the task's importance, define its scope a…
-
Content moderation and fraud detection rely on human-in-the-loop and ML patterns
Eugene Yan's article outlines five key patterns for building effective content moderation and fraud detection systems. These patterns emphasize collecting ground truth data through human input, augmenting this data, bre…
-
Mechanisms for Effective Technical Teams
Eugene Yan's article outlines several mechanisms to enhance the productivity and effectiveness of technical teams, particularly those involved in machine learning. Key practices include End-of-Week Debriefs (EOWDs) for …
-
Eugene Yan switches from Roam Research to Obsidian for note-taking
Eugene Yan details his migration from Roam Research to Obsidian, a process he found surprisingly straightforward and completed in under an hour. He outlines the steps involved, including downloading notes, organizing im…
-
Eugene Yan offers strategies for teams facing uncooperative dependency teams
Eugene Yan's article addresses the common challenge of inter-team dependencies, particularly when machine learning teams require assistance from data or infrastructure teams. The piece suggests moving beyond simple esca…
-
Eugene Yan reviews 2022, detailing career growth, writing goals, and investment thesis
Eugene Yan's 2022 review highlights personal and professional achievements, including writing 18 posts on technical topics like text-to-image and machine learning techniques. He was promoted from L5 to L6, focusing on M…
-
RecSys 2022 Keynote - Is the Juice Worth the Squeeze?
Eugene Yan delivered a keynote at the RecSys 2022 Workshop on Online Recommender Systems and User Modeling. His talk, titled "Online Recommender Systems: Is the juice worth the squeeze?", explored the trade-offs between…
-
Eugene Yan details robust testing strategies for data and ML pipelines
Eugene Yan's article explores methods for creating more resilient tests for data and machine learning pipelines. The author discusses why existing tests often fail even when new code is correct, attributing this to the …
-
Complexity bias favors complex ideas over simpler ones, despite benefits of simplicity
Eugene Yan argues that complexity is often favored over simplicity in technical fields due to a bias that equates complexity with effort, mastery, innovation, and more features. This bias leads to complex systems being …
-
Eugene Yan advocates for weekly 15-5 updates to boost team visibility and trust
Eugene Yan advocates for a weekly 15-5 update, a brief report designed to take 15 minutes to write and 5 minutes to read. This practice enhances team visibility by tracking work, outcomes, and blockers, thereby reducing…
-
Eugene Yan shares advice for effective onboarding in tech roles
Eugene Yan's article offers advice for effectively onboarding into new tech roles, emphasizing personal ownership of the process. He suggests proactively clarifying expectations, defining a 100-day plan, and building re…
-
Eugene Yan explains how to measure and mitigate position bias in recommendations
Position bias, where higher-ranked items receive more engagement regardless of relevance, poses a challenge for recommender systems. This bias can stem from user trust in algorithms, presentation effects, or a tendency …
-
Eugene Yan explains counterfactual evaluation for recommendation systems
Eugene Yan's article discusses the limitations of traditional offline evaluation for recommendation systems, arguing that they treat an interventional problem as observational. Current methods evaluate how well recommen…