Hello!

I'm Aftab Mallick

My works

Name: Aftab Mallick

Job Role: Agentic AI Engineer

Address: India

Skills

PYTHON 95%
LANGGRAPH 95%
FASTAPI 90%
MULTI-AGENT SYSTEMS 90%
KAFKA 85%
DOCKER 85%

About

About Me

Agentic AI Engineer and Lead Developer of TCS’s enterprise AI platform for 1,20,000+ users. Specialist in multi-agent systems, MCP, and high-throughput architectures.

  • Profile: Agentic AI
  • Domain: Multi-Agent Systems & AI Governance
  • Education: Bachelor of Technology in Computer Science
  • Language: English, Bengali, Hindi
  • Core Stack: Python, FastAPI, LangGraph, Agno, LiteLLM, MCP, Langfuse, Kafka
  • AI & Data: Multi-Agent Systems, RAG, Vector DBs (ChromaDB), Azure AI, Transformers
  • Infrastructure: Docker, Azure Service Bus, AWS, CI/CD, Event-driven Architecture

0 +   Projects completed

LinkedIn

Resume

Resume

AI/ML Engineer: Leveraging Data Science & Machine Learning to Drive Innovation and Impact

Experience


March 2025 - Present

AI Platform / Agentic AI Engineer

Tata Consultancy Services (TCS)
  • Led development of an enterprise Agentic AI platform (TCS Manufacturing AI Axis) for 1,20,000+ users, designing a provider-agnostic abstraction layer to facilitate seamless cross-cloud model interoperability while eliminating vendor lock-in.
  • Built a goal-based planner and multi-agent collaboration engine (Teams + Workflow services) converting natural language into executable systems; optimized prompt engineering pipelines and LiteLLM routing to reduce design time from 5 hrs to 5 mins.
  • Developed an event-driven workflow orchestration system using Kafka and Azure Service Bus, handling 2M+ events/day by implementing idempotent consumer patterns and dead-letter queueing with 99.5% reliability.
  • Designed a secure MCP framework and proxy for runtime tool injection; built centralized governance and policy engines with OAuth2 integration and RBAC protocols ensuring fine-grained access for all AI workloads.
  • Implemented observability via Langfuse with distributed tracing for agent systems, providing deep-dive bottleneck analysis and cost-tracking dashboards which reduced MTTR by 45%.


Education


2020-2024

Bachelor of Technology in Computer Science

MCKV Institute of Engineering

GPA: 8.96

Certifications

AI Engineer Production Track

Udemy - Deploy LLMs Agents at Scale

Python E2 Competency

Tata Consultancy Services

Download CV

Projects

Projects

Below are the sample Artificial Intelligence & Machine Learning projects on Python, FastAPI, LangGraph, Agno, LiteLLM, MCP, Langfuse, Kafka

Travel Planner Agentic Platform

Tech Stack: LangGraph, FastAPI, Serper, MongoDB, Multi-agent Orchestration

  • Built a stateful multi-agent system using DAG-based orchestration to coordinate Research and Planner agents with shared memory, state persistence, and parallelized sub-task execution.
  • Integrated Human-in-the-Loop (HITL) checkpoints for user validation before high-compute generation to optimize token usage, minimize hallucinations, and improve plan accuracy.
  • Developed a ReAct-based tool suite with web-grounding and weather-aware planning enabling context-rich, real-time itinerary generation.

View on GitHub

Adaptive Lifestyle Optimization Agent

Tech Stack: Agentic Reasoning, Multi-agent, LangGraph, MCP, Chroma

  • Designed a multi-agent lifestyle optimization system using LangGraph, architecting a state-aware cyclic graph to orchestrate Planner, Feedback, Adjuster, Memory, and NLU agents.
  • Implemented a hybrid memory architecture (MongoDB + ChromaDB semantic embeddings) enabling long-term behavioral reasoning.
  • Built real-world reasoning loops (plan → observe → analyze → replan) with dynamic tool usage via MCP.

View on GitHub

Artificial Intelligence
Machine Learning
Deep Learning
NLP

More projects on Github

I love to solve ML problems & uncover hidden data stories


GitHub

Contact

Contact Me

Below are the details to reach out to me!

Address

Bengaluru, India

Contact Number

+91 8100458986

Email Address

aftabmallick000@gmail.com

Download Resume

My Resume



Have a Question? Click Here


  • Find me on