back

AI Deep Research sucks at finding papers… So I built a better one

The search evolution: find my papers steps in

In an era of AI-driven research tools that often promise more than they deliver, a new contender has emerged that may actually fulfill the elusive promise of effective semantic paper discovery. Jay Alammar's latest project, "Find My Papers," tackles the shortcomings of current AI deep research functions with a refreshingly straightforward approach focused on quality results rather than flashy agent frameworks.

What sets this tool apart from the rest?

  • Purpose-built semantic search: The system currently indexes over 300,000 AI and machine learning papers from arXiv, enabling researchers to find conceptually similar papers even when terminology differs.

  • No hallucination framework: Unlike agentic systems like OpenAI's Deep Research or Perplexity, Find My Papers uses a custom retrieval pipeline without relying on multi-chain processes or neural-enabled networks, prioritizing accuracy over generative capabilities.

  • Rapid results: While deep research tools from major providers can take 3-10 minutes to compile their reports, this solution delivers comprehensive results in under two minutes.

  • Quality over synthetic summaries: Instead of generating potentially inaccurate 10,000-word reports, the system provides direct access to relevant papers with the ability to ask follow-up questions about specific findings.

The deeper insight: specialized retrieval beats generalized agents

The most compelling takeaway from this project isn't just that it outperforms current offerings, but why it does. By rejecting the industry trend toward generalized agentic systems that try to be everything to everyone, Find My Papers demonstrates the power of purpose-built systems with narrower scopes and deeper capabilities.

This approach matters because it represents a fundamental rethinking of how research tools should function. The current AI research landscape is dominated by "Swiss Army knife" models that promise to handle every task from writing emails to conducting doctoral research. But specialized tools with deliberate constraints often deliver superior results in their focused domains.

Consider how Bloomberg Terminal conquered financial data by doing one thing extremely well, rather than trying to be a general-purpose tool. Find My Papers appears to be following a similar philosophy for academic research.

Beyond the video: the growing specialized tools movement

What the video doesn't explore is how this tool fits into a broader emerging trend of specialized AI

Recent Videos

May 6, 2026

Hermes Agent Master Class

https://www.youtube.com/watch?v=R3YOGfTBcQg Welcome to the Hermes Agent Master Class — an 11-episode series taking you from zero to fully leveraging every feature of Nous Research's open-source agent. In this first episode, we install Hermes from scratch on a brand new machine with no prior skills or memory, walk through full configuration with OpenRouter, tour the most important CLI and slash commands, and run our first real task: a competitor research report on a custom children's book AI business idea. Every future episode will build on this fresh install so you can see the compounding value of the agent in real time....

Apr 29, 2026

Andrej Karpathy – Outsource your thinking, but you can’t outsource your understanding

https://www.youtube.com/watch?v=96jN2OCOfLs Here's what Andrej Karpathy just figured out that everyone else is still dancing around: we're not in an era of "better models." We're in a different era of computing altogether. And the difference between understanding that and not understanding it is the difference between being a vibe coder and being an agentic engineer. Last October, Karpathy had a realization. AI didn't stop being ChatGPT-adjacent. It fundamentally shifted. Agentic coherent workflows started to actually work. And he's spent the last three months living in side projects, VB coding, exploring what's actually possible. What he found is a framework that explains...

Mar 30, 2026

Andrej Karpathy on the Decade of Agents, the Limits of RL, and Why Education Is His Next Mission

A summary of key takeaways from Andrej Karpathy's conversation with Dwarkesh Patel In a wide-ranging conversation with Dwarkesh Patel, Andrej Karpathy — former head of AI at Tesla, founding member of OpenAI, and creator of some of the most popular AI educational content on the internet — shared his views on where AI is headed, what's still broken, and why he's now pouring his energy into education. Here are the key takeaways. "It's the Decade of Agents, Not the Year of Agents" Karpathy's now-famous quote is a direct pushback on industry hype. Early agents like Claude Code and Codex are...