frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: I reviewed about 300 academic papers of 2025 to write a book on startup

2•albertoasquer•1h ago
Hi HN, I’m the author of The 2025 Entrepreneurship Research Playbook.

I am an academic (SOAS University of London) who usually works in public policy. In 2023, I designed a "mini-MBA" for founders and realized a massive inefficiency: my students were dealing with problems that scholars study every day (liability of newness, signaling theory, dynamic capabilities, etc.), but they never saw the research because it was locked behind paywalls or written in dense academic-speak or just too recent to have added to the reading list!

I decided to bridge that gap. I performed a literature review and selected about 300 peer-reviewed entrepreneurship articles published in academic journals in 2025 and synthesized the findings into a practical guide.

Some non-obvious findings from the 2025 literature:

The "Ideator's Dilemma" in GenAI: Research shows GenAI reduces the cost of idea generation but creates a new cognitive burden: distinguishing between "hallucinated" viability and ideas that seem wrong only because they are novel/counter-intuitive (Chapter 3).

The "Passion" Paradox: While popular advice says "be passionate," recent studies (Jin et al., 2025) suggest that high-intensity passion in pitches actually increases perceived risk for investors unless balanced with specific "promotion-focused" language.

Signaling in Crowdfunding: "Linguistic hedging" (vague language) in campaign descriptions is now measurable as a negative signal, whereas specific disclosure of risks (not just strengths) increases trust/authenticity (Chapter 4).

The book is available here: https://entrepreneurshipscience.com/

I posted more contents in the linked website. I’m looking for feedback on which academic concepts feel most relevant to your actual daily grind as founders.

Happy to answer questions about the research!

(I hope the post is fine, I have just joined!)

On history and flattery

https://www.historians.org/perspectives-article/on-history-and-flattery/
1•hhs•3m ago•0 comments

Watch Claude Code debug WebGPU code without a GPU

https://www.youtube.com/watch?v=x791YvPIhFo
1•logicallee•4m ago•1 comments

PR Bro – a TUI that helps you decide what PR to review next

https://github.com/toniperic/pr-bro
1•toneric•7m ago•1 comments

The AI boom's surprising winners aren't even tech companies

https://www.axios.com/2026/02/05/ai-walmart-stock-wmt-consumer-goods
1•petethomas•8m ago•0 comments

CIA World Factbook Ends Publication After 6 Decades

https://www.nytimes.com/2026/02/05/us/world-factbook-cia.html
3•doener•9m ago•0 comments

A 16yo's mathematical derivation of Model Collapse (The Ainex Singularity)

https://zenodo.org/records/18434665
1•Mhh1430•10m ago•0 comments

Agents should not use Git anymore

https://ataraxy-labs.github.io/sem/
1•rohan2003•11m ago•1 comments

Show HN: Total Recall – write-gated memory for Claude Code

https://github.com/davegoldblatt/total-recall
1•davegoldblatt•12m ago•1 comments

Fusing communication and compute with new API and copy engine collective in NCCL

https://developer.nvidia.com/blog/fusing-communication-and-compute-with-new-device-api-and-copy-e...
1•teleforce•12m ago•0 comments

Homeland Security is trying to force tech to hand over data about Trump critics

https://www.yahoo.com/news/articles/homeland-security-trying-force-tech-183519291.html
5•c420•13m ago•0 comments

Kilo Pass by Kilo Code

https://kilo.ai/features/kilo-pass
1•satvikpendem•15m ago•0 comments

Simple Tutorial on How to Create a Hytale Server

https://hytalemultiplayer.io/blog/how-to-create-a-hytale-server/
1•kaddir•16m ago•0 comments

AMZN Q4 2025 Results and News Release

1•carlsborg•19m ago•0 comments

How does ChatGPT decide which websites to recommend?

1•nworley•19m ago•0 comments

Fairphone 6: cheaper, repairable and longer-lasting

https://www.theguardian.com/technology/2026/feb/04/fairphone-6-review-cheaper-repairable-longer-l...
4•montroser•19m ago•0 comments

China has seized Sony's television halo

https://www.ft.com/content/283ed0a2-48d1-4358-9021-decae9a61541
2•hhs•21m ago•0 comments

Amazon joins Big Tech AI spending spree

https://www.bbc.com/news/articles/c150e144we3o
2•Lyngbakr•22m ago•0 comments

Chris Hemsworth Is an L9 at Amazon, and I Have Questions

https://www.lastweekinaws.com/blog/chris-hemsworth-is-an-l9-at-amazon-and-i-have-questions/
2•mooreds•24m ago•2 comments

Pink noise reduces REM sleep and may harm sleep quality

https://www.pennmedicine.org/news/pink-noise-reduces-rem-sleep-and-may-harm-sleep-quality
4•hhs•24m ago•0 comments

Show HN: Watch LLMs Play NetHack

https://glyphbox.app/
1•kenforthewin•25m ago•1 comments

Moltbook – A social media for AI agents – Explained

https://diamantai.substack.com/p/moltbook-a-social-media-for-ai-agents
1•dskhatri•26m ago•0 comments

Show HN: SavvyLLM – Find the cheapest LLM for any task (800 models)

https://www.tryautropic.com/
1•savvyllm•26m ago•1 comments

Should You Buy a Newspaper or a Yacht?

https://www.theatlantic.com/newsletters/2026/02/bezos-washington-post-layoffs-yacht/685902/
2•petethomas•28m ago•0 comments

Writing an LLM from scratch, part 32c – Interventions: removing dropout

https://www.gilesthomas.com/2026/02/llm-from-scratch-32c-interventions-removing-dropout
1•gpjt•29m ago•0 comments

A 'crazy' dice proof leads to new understanding of a fundamental law of physics

https://phys.org/news/2026-02-crazy-dice-proof-fundamental-law.html
3•bikenaga•32m ago•1 comments

Ask HN: Anyone Using a Mac Studio for Local AI/LLM?

4•UmYeahNo•34m ago•2 comments

Amazon expects capex to hit $200B in 2026

https://www.cnbc.com/2026/02/05/amazon-amzn-q4-earnings-report-2025.html
2•mfiguiere•38m ago•0 comments

Show HN: Local task classifier and dispatcher on RTX 3080(No Ollama, raw Python)

https://github.com/resilientworkflowsentinel/resilient-workflow-sentinel
1•Shubham_Amb•38m ago•0 comments

Amazon Falls After Vowing to Spend $200B on AI This Year

https://www.bloomberg.com/news/articles/2026-02-05/amazon-boosts-spending-far-ahead-of-estimates-...
6•zerosizedweasle•38m ago•0 comments

Show HN: Cryptographic toolkit in a single HTML file

https://github.com/lownetjs/lownet-toolkit
1•BTMN•38m ago•0 comments