Pooja Maan profile photo 1 on Tal

Pooja Maan

AI Engineer @ IBM | Ex-Cisco | GenAI, LLMs & AI Specialist

Bengaluru

6+ years of work experience

IBM
Jul 20251 yr 3 mosPresent

Own development of AI capabilities across 5 automation loops within a broader platform of ~20–30 loops processing 100+ tickets/alarms per loop per day. Designed an AI-driven closed-loop workflow using LangGraph and Kafka to interpret incidents, enrich context, select remediation paths and route outcomes through automated action or human review, with runtime safeguards for controlled execution. Built new features across incident analysis, context enrichment, remediation and post-action validation; automated workflows currently handle ~80% of incidents, significantly reducing manual operational effort and turnaround time. Prototyped agent workflows in Langflow; built production AI services in Python, containerized them, and deployed to OpenShift through GitLab CI/CD. Mentored and onboarded 5 engineers, supporting their ramp-up on the agentic architecture, automation workflows and production engineering practices.
Cisco
Aug 20205 yrsJul 2025

Architected an AI-powered customer insights platform using Python/FastAPI and LLMs to analyze customer feedback and engineering support tickets, surface recurring failure patterns and contribute to a ~20% improvement in CSAT metrics. Built a RAG-based support simulation assistant using historical interaction logs, sentence embeddings and FAISS to generate realistic troubleshooting scenarios for engineering training. Engineered an end-to-end supervised ML solution to identify high-propensity accounts for Point of Sale attachments using historical enterprise deal data. Developed preprocessing, feature engineering, model training/evaluation and inference services; exposed predictions through FastAPI and deployed on GCP.
Cisco
Jan 20206 mosJun 2020

Automated technical configuration validation across Cisco CUBE environments using Python and REST APIs, reducing manual verification and audit-cycle overhead.

Pooja’s AI Stack

langchain
langchain
langgraph
langgraph
mcp
mcp
huggingface
huggingface
Milvus
Milvus
vLLM
vLLM

Education

National Institute of Technology, Tiruchirappalli

Master's degree, Data Analytics

National Institute of Technology, Tiruchirappalli

20182 yrs2020
Guru Gobind Singh Indraprastha University

Bachelor of Technology - BTech, Information Technology

Guru Gobind Singh Indraprastha University

20144 yrs2018

Join Pooja and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Pooja Maan on Tal

Headline

AI Engineer @ IBM | Ex-Cisco | GenAI, LLMs & AI Specialist

About

Senior AI Engineer with 6+ years of experience building and deploying production AI systems across IBM and Cisco. Hands-on expertise in Agentic AI, LLM applications, RAG, LangGraph orchestration, LLM evaluation and production safeguards, with strong software engineering foundations in Python, APIs, distributed systems and cloud-native deployment.

Location

Bengaluru

AI stack

langchain

langgraph

mcp

huggingface

Milvus

vLLM

Work experience

AI Engineer at IBM, Jul 2025 - Present · 1 yr 3 mos, Own development of AI capabilities across 5 automation loops within a broader platform of ~20–30 loops processing 100+ tickets/alarms per loop per day. Designed an AI-driven closed-loop workflow using LangGraph and Kafka to interpret incidents, enrich context, select remediation paths and route outcomes through automated action or human review, with runtime safeguards for controlled execution. Built new features across incident analysis, context enrichment, remediation and post-action validation; automated workflows currently handle ~80% of incidents, significantly reducing manual operational effort and turnaround time. Prototyped agent workflows in Langflow; built production AI services in Python, containerized them, and deployed to OpenShift through GitLab CI/CD. Mentored and onboarded 5 engineers, supporting their ramp-up on the agentic architecture, automation workflows and production engineering practices.

AI/ML Customer Experience at Cisco, Aug 2020 - Jul 2025 · 5 yrs, Architected an AI-powered customer insights platform using Python/FastAPI and LLMs to analyze customer feedback and engineering support tickets, surface recurring failure patterns and contribute to a ~20% improvement in CSAT metrics. Built a RAG-based support simulation assistant using historical interaction logs, sentence embeddings and FAISS to generate realistic troubleshooting scenarios for engineering training. Engineered an end-to-end supervised ML solution to identify high-propensity accounts for Point of Sale attachments using historical enterprise deal data. Developed preprocessing, feature engineering, model training/evaluation and inference services; exposed predictions through FastAPI and deployed on GCP.

CX Intern at Cisco, Jan 2020 - Jun 2020 · 6 mos, Automated technical configuration validation across Cisco CUBE environments using Python and REST APIs, reducing manual verification and audit-cycle overhead.

Education

Master's degree, Data Analytics, National Institute of Technology, Tiruchirappalli

Bachelor of Technology - BTech, Information Technology, Guru Gobind Singh Indraprastha University

Pooja Maan - AI Engineer in Bengaluru | Tal