Komal C. profile photo 1 on Tal

Komal C.

Data Science @ProArch | Hansraj, DU

Haldwani

3+ years of work experience

ProArch
Feb 20267 mosPresent

ProArch
Jul 20258 mosFeb 2026

Supply Chain Forecasting and allocation engine
ProArch
Dec 20248 mosJul 2025

• Microsoft Fabric Data Engineering: Built end to end pipelines using Medallion Architecture to get data from Hubspot and Autotask to facilitate BI reporting and Agentic AI workings on top of it • Email Multi-Label Classification: Trained and deployed transformer models (BERT, RoBERTa, DistilBERT) to classify synthetically generated promotional emails into Scan, Rebate, or Lump Sum categories. • AI Avatar: Created an interactive voice+video avatar using open-source models (Coqui-TTS, SadTalker). Used Azure Speech Studio, and ElevenLabs to build a Flask chatbot enabling real-time voice cloning and user voice interaction with avatar. • AI-Driven SQA Tool: Developed a Generative AI tool to auto-generate test cases, test data, metadata, and Selenium scripts from SRS documents, UI screenshots, and user stories; integrated generated tests into Azure DevOps.
Aays
Jul 20231 yr 2 mosAug 2024

Gurugram, Haryana, India

• OpenAI LLM Fine-tuning: Fine-tuned GPT-3.5 using OpenAI API to extract KPIs from PyTesseract OCR text in non-scannable PDFs; generated training data for custom LLM models. • Zero-Shot Classification: Implemented zero-shot classification models with Hugging Face transformers to segment customer service queries, revealing customer behavior insights. • Power BI Dashboards: Delivered advanced dashboards featuring complex DAX and dynamic visualizations for finance and innovation projects. • Client Collaboration: Translated wireframes into dashboards and scalable solutions across Excel, Azure, and Databricks; praised for delivering innovative insights even under tight deadlines. • RAG System Development: Developed a Retrieval-Augmented Generation (RAG) component within a React + Vite app (ReactFlow) using a mind map approach to enhance LLM contextual responses.
Indian Institute of Technology, Madras
Aug 20222 mosSep 2022

Lead Python Mentor: Organised coding sessions everyday and helped students build coding logic, debug code and gave hints as per need. Worked with two sub-mentors and a student group of 16.
Enactus Hansraj
Jan 20216 mosJun 2021

Delhi, India

GitHub
@komalchamyal

Komal’s AI Stack

antigravity
antigravity
codex
codex

Education

Hansraj College

Bachelor of Science - BS, Computer Science

Hansraj College

20203 yrs2023
Indian Institute of Technology, Madras

Diploma in Programming

Indian Institute of Technology, Madras

20212 yrs2023

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 C. on Tal

Headline

Data Science @ProArch | Hansraj, DU

About

Passionate about turning data into actionable insights and measurable impact, I enjoy solving complex problems, building scalable models, and translating analytics into clear business value. Always learning, always improving, and always curious about the “why” behind the numbers.

Location

Haldwani

AI stack

antigravity

codex

Work experience

Data Scientist at ProArch, Feb 2026 - Present · 7 mos

Associate Data Scientist at ProArch, Jul 2025 - Feb 2026 · 8 mos, Supply Chain Forecasting and allocation engine

Junior Data Scientist at ProArch, Dec 2024 - Jul 2025 · 8 mos, • Microsoft Fabric Data Engineering: Built end to end pipelines using Medallion Architecture to get data from Hubspot and Autotask to facilitate BI reporting and Agentic AI workings on top of it • Email Multi-Label Classification: Trained and deployed transformer models (BERT, RoBERTa, DistilBERT) to classify synthetically generated promotional emails into Scan, Rebate, or Lump Sum categories. • AI Avatar: Created an interactive voice+video avatar using open-source models (Coqui-TTS, SadTalker). Used Azure Speech Studio, and ElevenLabs to build a Flask chatbot enabling real-time voice cloning and user voice interaction with avatar. • AI-Driven SQA Tool: Developed a Generative AI tool to auto-generate test cases, test data, metadata, and Selenium scripts from SRS documents, UI screenshots, and user stories; integrated generated tests into Azure DevOps.

Associate Data Scientist at Aays, Jul 2023 - Aug 2024 · 1 yr 2 mos, Gurugram, Haryana, India, • OpenAI LLM Fine-tuning: Fine-tuned GPT-3.5 using OpenAI API to extract KPIs from PyTesseract OCR text in non-scannable PDFs; generated training data for custom LLM models. • Zero-Shot Classification: Implemented zero-shot classification models with Hugging Face transformers to segment customer service queries, revealing customer behavior insights. • Power BI Dashboards: Delivered advanced dashboards featuring complex DAX and dynamic visualizations for finance and innovation projects. • Client Collaboration: Translated wireframes into dashboards and scalable solutions across Excel, Azure, and Databricks; praised for delivering innovative insights even under tight deadlines. • RAG System Development: Developed a Retrieval-Augmented Generation (RAG) component within a React + Vite app (ReactFlow) using a mind map approach to enhance LLM contextual responses.

Python Student Mentor at Indian Institute of Technology, Madras, Aug 2022 - Sep 2022 · 2 mos, Lead Python Mentor: Organised coding sessions everyday and helped students build coding logic, debug code and gave hints as per need. Worked with two sub-mentors and a student group of 16.

Coordinator at Enactus Hansraj, Jan 2021 - Jun 2021 · 6 mos, Delhi, India

Education

Bachelor of Science - BS, Computer Science, Hansraj College

Diploma in Programming, Indian Institute of Technology, Madras

Public links

github: https://github.com/komalchamyal