corpus.blog
56,966 blogs · [ { "id": "01a0e495-a10a-7034-8f80-0f7816331975", "title": "Sandboxing Megasthenes", "url": "http://blog.nilenso.com/blog/2026/06/25/sandboxing-ai-agents/", "published_at": "2026-06-25T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f7816e2081f", "title": "Building the systems that build the software", "url": "http://blog.nilenso.com/blog/2026/06/12/building-the-systems-that-build-the-software/", "published_at": "2026-06-12T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f78179cc825", "title": "Our approach to evals with megasthenes", "url": "http://blog.nilenso.com/blog/2026/06/10/our-approach-to-evals-with-megasthenes/", "published_at": "2026-06-10T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f7817f6de6f", "title": "Training a small model to write better OCaml with RLVR and GRPO", "url": "http://blog.nilenso.com/blog/2026/05/18/training-a-small-model-to-write-better-ocaml-with-rlvr-and-grpo/", "published_at": "2026-05-18T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f7818980d5f", "title": "Evals before prompts: building an LLM OCR for KYC", "url": "http://blog.nilenso.com/blog/2026/05/18/evals-before-prompts-building-an-llm-ocr-for-kyc/", "published_at": "2026-05-18T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f7818d53e3b", "title": "Temporal anti-pattern: Don't treat expected failures as exceptions", "url": "http://blog.nilenso.com/blog/2026/04/30/business-errors-are-outcomes-not-exceptions/", "published_at": "2026-04-30T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781928fa55", "title": "Trajectory shapes", "url": "http://blog.nilenso.com/blog/2026/04/20/trajectory-shapes/", "published_at": "2026-04-20T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781a0db331", "title": "Checking my model vibes against SWE-Bench Pro", "url": "http://blog.nilenso.com/blog/2026/04/08/checking-my-model-vibes-against-swe-bench-pro/", "published_at": "2026-04-08T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781a8624cc", "title": "Feature Platforms: The Underrated Infrastructure Layer Behind Fast ML Teams", "url": "http://blog.nilenso.com/blog/2026/02/19/feature-platforms/", "published_at": "2026-02-19T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781b1e4716", "title": "Engineering Maturity is all you need", "url": "http://blog.nilenso.com/blog/2026/02/16/engineering-maturity-is-all-you-need/", "published_at": "2026-02-16T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781b903f72", "title": "Weird system prompt artefacts", "url": "http://blog.nilenso.com/blog/2026/02/12/weird-system-prompt-artefacts/", "published_at": "2026-02-12T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781c58c680", "title": "How System Prompts Reveal Model Biases", "url": "http://blog.nilenso.com/blog/2026/02/12/how-system-prompts-reveal-model-biases/", "published_at": "2026-02-12T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781c973ee2", "title": "Codex CLI vs Claude Code on autonomy", "url": "http://blog.nilenso.com/blog/2026/02/12/codex-cli-vs-claude-code-on-autonomy/", "published_at": "2026-02-12T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781d78e27d", "title": "How System Prompts Define Agent Behavior", "url": "http://blog.nilenso.com/blog/2026/02/10/how-system-prompts-define-agent-behaviiour/", "published_at": "2026-02-10T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781e009724", "title": "Reinforcement Learning with GRPO", "url": "http://blog.nilenso.com/blog/2026/01/28/reinforcement-learning-with-grpo/", "published_at": "2026-01-28T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781e721e4f", "title": "How the Lobsters front page works", "url": "http://blog.nilenso.com/blog/2026/01/20/lobsters-front-page/", "published_at": "2026-01-20T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781e7ad944", "title": "Minimum Viable Benchmark", "url": "http://blog.nilenso.com/blog/2025/11/28/minimum-viable-benchmark/", "published_at": "2025-11-28T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781e96ec22", "title": "How to work with Product: Taste and Adjust", "url": "http://blog.nilenso.com/blog/2025/11/26/how-to-work-with-product-taste-and-adjust/", "published_at": "2025-11-26T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781eb22e36", "title": "How to work with Product: To What Port Do You Sail?", "url": "http://blog.nilenso.com/blog/2025/11/21/how-to-work-with-product-towards-what-port-do-you-sail/", "published_at": "2025-11-21T00:00:00+00:00" }, { "id": "01a0e495-a10a-7034-8f80-0f781ed90efa", "title": "How to work with Product: At the Tea Table", "url": "http://blog.nilenso.com/blog/2025/11/18/how-to-work-with-product-at-the-tea-table/", "published_at": "2025-11-18T00:00:00+00:00" } ] posts Claim your blog
Back to blog.nilenso.com
Blog · corpus.blog/blogs/blog.nilenso.com/posts

blog.nilenso.com

blog.nilenso.com

2026

Sandboxing Megasthenesoriginal ↗
25 Jun 2026
Building the systems that build the softwareoriginal ↗
12 Jun 2026
Our approach to evals with megasthenesoriginal ↗
10 Jun 2026
Training a small model to write better OCaml with RLVR and GRPOoriginal ↗
18 May 2026
Evals before prompts: building an LLM OCR for KYCoriginal ↗
18 May 2026
Temporal anti-pattern: Don't treat expected failures as exceptionsoriginal ↗
30 Apr 2026
Trajectory shapesoriginal ↗
20 Apr 2026
Checking my model vibes against SWE-Bench Prooriginal ↗
8 Apr 2026
Feature Platforms: The Underrated Infrastructure Layer Behind Fast ML Teamsoriginal ↗
19 Feb 2026
Engineering Maturity is all you needoriginal ↗
16 Feb 2026
Weird system prompt artefactsoriginal ↗
12 Feb 2026
How System Prompts Reveal Model Biasesoriginal ↗
12 Feb 2026
Codex CLI vs Claude Code on autonomyoriginal ↗
12 Feb 2026
How System Prompts Define Agent Behaviororiginal ↗
10 Feb 2026
Reinforcement Learning with GRPOoriginal ↗
28 Jan 2026
How the Lobsters front page worksoriginal ↗
20 Jan 2026

2025

Minimum Viable Benchmarkoriginal ↗
28 Nov 2025
How to work with Product: Taste and Adjustoriginal ↗
26 Nov 2025
How to work with Product: To What Port Do You Sail?original ↗
21 Nov 2025
How to work with Product: At the Tea Tableoriginal ↗
18 Nov 2025