Ajay Pravin Mahale
GitHub

Ajay Pravin Mahale

@designer-coderajay

View GitHub Profile

About

AI/ML Researcher & Engineer | Mechanistic Interpretability | LLM Safety | NLP | Azure ML & MLOps | PyTorch | TransformerLens

Location Germany
Followers 6
Following 5
Public Repositories 19
Expertise

Skills & Technologies

Python TypeScript
GitHub Work

Projects & Repositories

Recent public projects and repositories from this profile.

Circuit-level-interpretability-evidence-does-not-survive-defensible-analytic-variation

Python

Circuit-level interpretability evidence does not survive defensible analytic variation. Pre-registered multiverse over 15,840 specifications; the derived EU AI Act Annex IV claim flips across 73.2% of specification pairs. arXiv:2608.13754

⭐ 1 Forks: 0

enterprise-agentic-ai-platform

Python

Production-grade multi-agent AI platform with LangGraph v0.3, MCP servers, hybrid RAG (Qdrant + LlamaIndex), FastAPI WebSocket streaming, and Next.js real-time UI. Deployed on AWS EKS via Terraform.

⭐ 5 Forks: 3

Portfolio

TypeScript

Portfolio

⭐ 1 Forks: 0
TypeScript View Project

designer-coderajay

Config files for my GitHub profile.

⭐ 0 Forks: 0
Repository View Project

Causal-Attribution-for-Agentic-Decisions

Python

Post-hoc causal attribution over LLM-agent trajectories. Two total effects that both fail, a closed form for coupling once contexts diverge, and a traceability specification read against Regulation (EU) 2024/1689. Pre-registered experiment published in full and not run.

⭐ 1 Forks: 0

glassbox-mech

Python

Open-source EU AI Act Annex IV documentation toolkit. Mechanistic interpretability + circuit discovery for transformers. One function call generates a structured, hash-chained evidence package.

⭐ 3 Forks: 0