Skip to content
View Aftabmallick's full-sized avatar
2 followers · 1 following

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Sharkx2

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Sharkx2

Highlights

Block or report Aftabmallick

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.

Content in all repositories owned by your account will be closed.
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
Aftabmallick/README.md

Aftab Mallick

AI Software Engineer | Python, Distributed Systems & Agentic Workflows
Hyderabad, India • LinkedIn • aftabmallick000@gmail.com

📌 Overview

AI Software Engineer with enterprise experience building production-grade LLM applications, stateful multi-agent workflows, and distributed backend systems. Focused on building deterministic AI architectures, multi-tenant retrieval pipelines, and robust evaluation gates.

  • Current Work: Enterprise agentic platforms, tool integration via Model Context Protocol (MCP), and asynchronous pipeline orchestration.
  • Core Stack: Python (Asyncio, FastAPI), LangGraph, LiteLLM, Celery, Kafka, Redis, PostgreSQL (RLS), Qdrant.
  • Reliability & Telemetry: Evaluation harnesses, distributed tracing (OpenTelemetry, Langfuse), and guardrail enforcement.

🚀 Featured Systems

⚡ TitanRAG

Production-grade Multi-Tenant Multimodal RAG Platform

  • Strict tenant data isolation via PostgreSQL Row-Level Security (RLS) and partitioned Qdrant collections.
  • Decoupled worker architecture using a transactional outbox pattern with Celery for heavy OCR and embedding pipelines.
  • Hybrid search (dense vectors + BM25) with cross-encoder re-ranking and sub-80ms Redis semantic caching.

🛡️ Payment Collection Agent

Deterministic Agentic State Machine with Regulatory Compliance

  • Stateful conversational flow driven by LangGraph state machines for negotiation, verification, and payment collection.
  • Strict PCI-DSS & PII sanitization using automated regex/NER scrubbing, Luhn validation, and encrypted tokenization.
  • State persistence via MongoDB checkpoints with automated fallback escalation to human operators.

🛠️ Technical Capabilities

  • Languages & Frameworks: Python, FastAPI, SQL, Asynchronous Programming, Pydantic
  • LLM & Agent Systems: LangGraph, Multi-Agent Coordination, Model Context Protocol (MCP), LiteLLM, RAG Architecture, Tool Calling
  • Data & Messaging: Apache Kafka, Celery, Azure Service Bus, Redis, PostgreSQL (RLS), Qdrant, ClickHouse
  • Observability & Infrastructure: OpenTelemetry, Langfuse, Docker, CI/CD (GitHub Actions), Git

Pinned Loading

  1. titanrag titanrag Public

    TitanRAG - An open-source RAG infrastructure for building scalable, multi-tenant and multimodal AI applications.

    Python 3

  2. horus horus Public

    Horus - Observability and distributed tracing platform for autonomous AI agents and LLM applications.

    Python