Eeshan S.
--:--
⛅ --°F Seattle

Concepts

I build useful things with AI, usually data science and gamification. The idea is to keep going back and de-vibing the prototypes. Click a concept to filter the grid by topic.

Projects

A/B Testing Memory Game
A live A/B test on game difficulty and player behavior
Astro / React / TailwindPostHogSupabasePython / statsmodels
Live
Try it Source
How I Prompt 2025
Year-in-review visualization of your Claude conversations
PythonClaude AIData Visualization
Live
Try it Source
Local LLM Bench
Benchmark local LLMs on structured data extraction accuracy
LocalAIMLXOllamaPython
Live
Try it Source
Sidequests
A hands-off dashboard for every local side project
Next.js / ReactNode CLIPrisma / libSQLClaude Code + Codex
Live
Try it Source

Writing | The Asymptotic
How I De-Vibed a Vibe-Coded NLP App
Apr 6, 2026 LanguageEssays
Agentic Coding for Non-Vibe Coders
Mar 8, 2026 Essays
The Dopamine Trap of AI Coding
Feb 23, 2026 Essays
Local AI matters, but it's not ready yet
Jan 25, 2026 AI on Your Own MachineEssays
What 4,887 AI prompts say about how we talk to machines
Jan 2, 2026 LanguageData Visualization
From game design to p-values: Anatomy of a real A/B experiment
Dec 21, 2025 Experimentation

GitHub

@eeshansrivastava89
1,087 contributions in the last 6 months
643 commits in 10 repositories
datascienceapps 266 commits howiprompt 164 commits sidequests 117 commits ab-simulator 27 commits brewfolio 26 commits
Created 10 repositories
brewfolio Astro hermes-workspace TypeScript
Opened 133 issues
datascienceapps Phase 5: GitHub storage mode for Keystatic (edit from any device) local-llm-bench References hermes-workspace Onboarding doesn't mention API_SERVER_ENABLED=true requirement
Opened 1 pull request
hermes-workspace fix: auth-check endpoint bypasses gateway port auto-detection

Analysis
A/B Test: Pineapple Finder Difficulty
Dec 24, 2025
from The Asymptotic

Writing archive

Essays on experimentation, local AI, prompting, and product thinking. The archive is organized like a compact editorial index: one featured read, then chronological shelves.

Latest

How I De-Vibed a Vibe-Coded NLP App

Earlier in the year, I used AI to build me a prompt analyzer app in 20 mins. It took me weeks to wade through the slop and apply the right methods to make it work correctly.

April 6, 2026 LanguageEssays Open article →

2026

5 pieces
How I De-Vibed a Vibe-Coded NLP App
Earlier in the year, I used AI to build me a prompt analyzer app in 20 mins. It took me weeks to wade through the slop and apply the right methods to make it work correctly.
Apr 6, 2026 LanguageEssays
Agentic Coding for Non-Vibe Coders
A workflow for building real things (not demos) with AI coding agents
Mar 8, 2026 Essays
The Dopamine Trap of AI Coding
The victory of solving a hard problem was sweeter when things used to take time. Now, getting a dopamine hit from watching an agent do all the work, is not the same at all.
Feb 23, 2026 Essays
Local AI matters, but it's not ready yet
Frontier models are extraordinary. On-device alternatives are not there yet for the average person. But the privacy and cost math is shifting fast.
Jan 25, 2026 AI on Your Own MachineEssays
What 4,887 AI prompts say about how we talk to machines
I analyzed a year of my Claude conversations. The patterns were more revealing than the content itself.
Jan 2, 2026 LanguageData Visualization

2025

1 pieces
From game design to p-values: Anatomy of a real A/B experiment
Most data scientists inherit someone else's pipeline. I built one end-to-end — product design, instrumentation, statistics — to show how it's done at the highest levels.
Dec 21, 2025 Experimentation
Subscribe on Substack