Skip to content
Atlas profile

Pranav

@pranavsinghpatilVerified GitHub profile

Investing my hours in LLMs

AI-Atlas score

Specialist
0100
Strongest
AI Agents
74.4/100 domain score
Rank
4th=
of 86
India
1st
of 12
  • IndiaLocation
  • 15 Nov 2026Mapped
  • 30 / 42AI projects
01Skills matrix

AI Skills Matrix

AI Skills Matrix scores, 0 to 100
Machine Learning67
Computer Vision22
Data Science52
NLP & LLMs70
AI Agents74
MLOps23

Each node is one domain, scored only from evidence in these public repositories. Select one to see why.

Deep, repeated, well-engineered work in this domain.

02Evaluated projects6 of 30

42 public repositories → 30 with AI work, evaluated. The other 12 (repositories without AI code, forks and empty repositories) are not scored and have no effect on the result. How repositories are selected

Ranked by contribution

Ordered by how much each project adds to the Atlas score, not by its own score. A project with evidence in several domains can rank above one with a slightly higher project score. Choose a domain to focus on the projects with meaningful evidence there.

01
58.1

Implementation

17% of Atlas Score

Educational Implementation· Jupyter Notebook

This repository contains code from the book 'Build a Large Language Model From Scratch', including implementations of GPT-like architectures from scratch, training scripts for Hugging Face models, and user interfaces for model interaction. It serves as an educational resource for learning LLMs through practical implementation.

NLP & LLMs0.95Machine Learning0.85Data Science0.40MLOps0.20
02
78.6

Expert

11% of Atlas Score

AI Developer Tool· Python

The Harness-Agent repository provides a dual-system architecture for autonomous hardware reliability investigation, featuring a custom virtual silicon hardware sandbox and an autonomous Bayesian investigator that plans experiments to test and debug agent behavior under realistic hardware faults.

AI Agents0.90
agent-harnessagentic-aiai-agentsai-safetyautonomous-agents
03
63.1

Advanced

11% of Atlas Score

Agent System· Python

The Autonomous Task Worker is a Python-based prototype for an autonomous AI agent that performs invoice intake in a simulated company environment. It uses a planner (with LLM and offline fallback) to generate a validated plan, executes typed tools against a local SQLite database, and verifies outcomes before reporting completion.

AI Agents0.90NLP & LLMs0.80Data Science0.20
04
62.9

Advanced

9% of Atlas Score

AI Developer Tool· Python

Model Anatomy is a full-stack application for inspecting and visualizing the internal architecture of Hugging Face models. It provides a CLI, web interface, and API to analyze model checkpoints and generate evidence-based architecture graphs.

Machine Learning0.85NLP & LLMs0.65
05
60.7

Advanced

7% of Atlas Score

Dataset / Data Pipeline· JavaScript

OpenEvents is a full-stack application that aggregates leisure events from multiple web sources using Bright Data scrapers, validates, normalizes, and deduplicates the data, then serves it via a FastAPI backend and React frontend.

Data Science0.70NLP & LLMs0.40
06
58.7

Advanced

6% of Atlas Score

Educational Implementation· Python

The repository implements character, word, and BPE tokenizers from scratch, accompanied by detailed explanations, benchmarks, and tests to teach tokenization fundamentals for LLMs.

NLP & LLMs0.90Machine Learning0.60Data Science0.20