Skip to content
View ama228's full-sized avatar
0 followers · 1 following
  • Open to remote work
  • India

Block or report ama228

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ama228/README.md

Hi 👋 Welcome

I'm Amar

I'm a Python developer and AI/ML engineer focused on building practical tools for LLM evaluation, RAG systems, prompt engineering, and code analysis. I like solving hard problems and shipping clean, well-tested code.

🚀 AI/ML Projects

Project What it does Tests
rag-pipeline RAG system with chunking, vector store, hybrid search (vector + keyword), and retrieval 44 ✅
llm-eval LLM evaluation toolkit — ROUGE-L, BLEU, fuzzy match, code correctness, RLHF pairwise comparison 38 ✅
ai-code-reviewer AST-based Python code reviewer with complexity analysis and optional LLM integration 22 ✅
promptlab Prompt template engine with conditionals, runner with retries, and automated testing framework 27 ✅

🛠️ Python Libraries

Project What it does Tests
smart-config-parser Config string parser with quoted values, nested keys, ranges, and validation 34 ✅
pypool Thread-safe connection pool with overflow, timeouts, and context manager support 24 ✅
timedeltaparser Parse human-readable durations like 1h 30m 15s into timedelta objects 32 ✅
pytaskrunner Task runner with dependency resolution and cycle detection 16 ✅
cachelimit HTTP response cache with size budgets, TTL, and LRU eviction 31 ✅

💡 What I Work On

  • RAG Pipelines — chunking strategies, vector search, hybrid retrieval, context window optimization
  • LLM Evaluation — metrics (ROUGE-L, BLEU, code correctness), RLHF-style pairwise comparison, automated eval pipelines
  • Prompt Engineering — template systems, prompt testing frameworks, message formatting for chat models
  • Code Analysis — AST-based static analysis, pattern detection, complexity scoring
  • Backend Systems — connection pooling, caching, task scheduling, CLI tooling

🔧 Technologies I Use

Python   pytest   Git   GitHub   SQLAlchemy   FastAPI   Docker   Linux   VS Code

🤝 Open Source Contributions

Amar's GitHub stats

Building tools that make AI development faster and more reliable.

👉 Currently open for remote contract work in AI training, LLM evaluation, code generation, and backend engineering. Let's connect.

Pinned Loading

  1. ai-code-reviewer ai-code-reviewer Public

    AST-based Python code reviewer with pattern detection and optional LLM integration

    Python

  2. llm-eval llm-eval Public

    LLM evaluation toolkit - ROUGE-L, fuzzy match, code correctness metrics with RLHF-style pairwise comparison

    Python

  3. promptlab promptlab Public

    Prompt engineering toolkit - templates with variables/conditionals, runner with retries, and automated testing framework

    Python

  4. pypool pypool Public

    Thread-safe connection pool with overflow, configurable timeouts and stats tracking

    Python

  5. rag-pipeline rag-pipeline Public

    Lightweight RAG pipeline library - chunking, vector store, retrieval, and generation with pluggable components

    Python

  6. smart-config-parser smart-config-parser Public

    Flexible config string parser with key:value pairs, quoted values, ranges and validation

    Python