KOMAL . profile photo 1 on Tal

KOMAL .

Lead ML Engineer @ airtel | MTech in Artificial Intelligence

New Delhi

5+ years of work experience

airtel
Feb 20251 yr 7 mosPresent

Gurugram, Haryana, India

- Led development of an AI-powered voice assistant for automated broadband troubleshooting using LangGraph, LLMs, and real-time speech pipelines. - Built and optimized Hindi/Hinglish speech recognition systems using Whisper for noisy call environments. - Designed scalable STT/TTS microservices with FastAPI and WebSockets for high-volume real-time traffic. - Implemented LLM-based workflows for decision-making, OTP extraction, and appointment rescheduling. - Developed document-grounded RAG systems using LlamaIndex, Elasticsearch, and vector databases. - Built end-to-end 5G usage forecasting systems using ARIMA, Prophet, and LSTM models. - Designed preprocessing and feature engineering pipelines for large-scale time series data. - Integrated structured logging, monitoring, and observability for production stability.
Zee Media Corporation Limited
Jun 20249 mosFeb 2025

Uttar Pradesh, India

• Designed and deployed scalable AI-based content moderation systems using FastAPI, Kafka, GPT models, and OpenCV for real-time audio/video analysis and regulatory compliance. • Built AI-driven video processing pipelines for frame change detection and consumer product identification using OpenCV and OpenAI. • Developed automated tools for generating headlines, highlights, and keywords using OpenAI models for multilingual content optimization. • Built an AI-powered document processing chatbot using GPT, Vision Transformers, OCR, and SentenceTransformers, integrated with OpenSearch and FastAPI, improving user engagement by 40%. • Applied advanced NLP models (GPT-4, Hugging Face, BERT) for text classification and multilingual content processing using Google Translation API.
Zee Media Corporation Limited
Feb 20222 yrs 5 mosJun 2024

Noida, Uttar Pradesh, India

• Created SEO dashboards integrating Comscore APIs, reducing reporting time by 80%, deployed with Docker and CI/CD on GitHub. • Built automated data pipelines integrating YouTube, Google Analytics, Comscore, and Tubular for centralized reporting. • Created interactive Power BI dashboards for business intelligence and performance analysis. • Performed large-scale data cleaning, feature engineering, and modeling for content analytics. • Supported editorial and business teams with data-driven insights and performance reports.
VyomNetra
Jul 20216 mosDec 2021

Nodia, UP

• Developed touchless visitor management system with AI and facial recognition, enhancing facility security, and providing data for traffic analysis. • Created IoT-based smart washroom system with ammonia sensor, enabling real-time cleanliness monitoring and predictive maintenance for improved facility hygiene. • Demonstrated proficiency in Python, machine learning techniques, and IoT technology throughout project lifecycles. • Offered data-driven recommendations for resource allocation and process optimization, supporting the startup's growth initiatives.

KOMAL’s AI Stack

opencode
opencode
claudecode
claudecode

Education

Indira Gandhi Delhi Technical University for Women

Master of Technology - MTech, Artificial Intelligence

Indira Gandhi Delhi Technical University for Women

20192 yrs2021
Jamia hamdard

Bachelor's degree, Electronics and Communications Engineering

Jamia hamdard

20154 yrs2019

Join KOMAL and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

KOMAL . on Tal

Headline

Lead ML Engineer @ airtel | MTech in Artificial Intelligence

About

As a Lead Machine Learning Engineer at airtel, I focus on designing and deploying AI-powered voice assistants and real-time speech pipelines for automated broadband troubleshooting. My work incorporates technologies such as LangGraph, LLMs, Whisper, and FastAPI to optimize Hindi/Hinglish speech recognition, build scalable STT/TTS microservices, and implement LLM-based workflows for decision-making and task automation. Additionally, I contribute to the development of large-scale 5G usage forecasting systems using ARIMA, Prophet, and LSTM models, alongside preprocessing and feature engineering for time series data. I hold an MTech in Artificial Intelligence from Indira Gandhi Delhi Technical University for Women and bring a strong foundation in AI-driven solutions to my role. My expertise spans Generative AI, RAG systems, and speech recognition technologies, with a focus on creating scalable, production-ready AI systems that address complex business challenges.

Location

New Delhi

AI stack

opencode

claudecode

Work experience

Lead ML Engineer at airtel, Feb 2025 - Present · 1 yr 7 mos, Gurugram, Haryana, India, - Led development of an AI-powered voice assistant for automated broadband troubleshooting using LangGraph, LLMs, and real-time speech pipelines. - Built and optimized Hindi/Hinglish speech recognition systems using Whisper for noisy call environments. - Designed scalable STT/TTS microservices with FastAPI and WebSockets for high-volume real-time traffic. - Implemented LLM-based workflows for decision-making, OTP extraction, and appointment rescheduling. - Developed document-grounded RAG systems using LlamaIndex, Elasticsearch, and vector databases. - Built end-to-end 5G usage forecasting systems using ARIMA, Prophet, and LSTM models. - Designed preprocessing and feature engineering pipelines for large-scale time series data. - Integrated structured logging, monitoring, and observability for production stability.

Machine Learning Engineer at Zee Media Corporation Limited, Jun 2024 - Feb 2025 · 9 mos, Uttar Pradesh, India, • Designed and deployed scalable AI-based content moderation systems using FastAPI, Kafka, GPT models, and OpenCV for real-time audio/video analysis and regulatory compliance. • Built AI-driven video processing pipelines for frame change detection and consumer product identification using OpenCV and OpenAI. • Developed automated tools for generating headlines, highlights, and keywords using OpenAI models for multilingual content optimization. • Built an AI-powered document processing chatbot using GPT, Vision Transformers, OCR, and SentenceTransformers, integrated with OpenSearch and FastAPI, improving user engagement by 40%. • Applied advanced NLP models (GPT-4, Hugging Face, BERT) for text classification and multilingual content processing using Google Translation API.

Data Analyst AI at Zee Media Corporation Limited, Feb 2022 - Jun 2024 · 2 yrs 5 mos, Noida, Uttar Pradesh, India, • Created SEO dashboards integrating Comscore APIs, reducing reporting time by 80%, deployed with Docker and CI/CD on GitHub. • Built automated data pipelines integrating YouTube, Google Analytics, Comscore, and Tubular for centralized reporting. • Created interactive Power BI dashboards for business intelligence and performance analysis. • Performed large-scale data cleaning, feature engineering, and modeling for content analytics. • Supported editorial and business teams with data-driven insights and performance reports.

Artificial Intelligence Engineer at VyomNetra, Jul 2021 - Dec 2021 · 6 mos, Nodia, UP, • Developed touchless visitor management system with AI and facial recognition, enhancing facility security, and providing data for traffic analysis. • Created IoT-based smart washroom system with ammonia sensor, enabling real-time cleanliness monitoring and predictive maintenance for improved facility hygiene. • Demonstrated proficiency in Python, machine learning techniques, and IoT technology throughout project lifecycles. • Offered data-driven recommendations for resource allocation and process optimization, supporting the startup's growth initiatives.

Education

Master of Technology - MTech, Artificial Intelligence, Indira Gandhi Delhi Technical University for Women

Bachelor's degree, Electronics and Communications Engineering, Jamia hamdard