Data Science & AI/ML Architect / Engineer · 11+ YOE

NageshCH

Agentic AI · LLM Engineering · Data Engineering · Cloud · Enterprise AI Strategy

11+ years building production AI systems — multi-agent orchestration, RAG pipelines, LLM evaluation, and enterprise-scale data engineering. Based in Hyderabad.

View Work Get in Touch

$9M+

Business Impact

11+

Years Experience

$1.5M

Largest AI Deal

93%

BERT Accuracy

90%

Time Reduction

60%

MTTR Reduction

About Me

I'm a Data Science & AI/ML Architect / Engineer with 11+ years shipping production-grade GenAI, Agentic AI & ML systems across CPG, Telecom, and Chemical/Materials industries.

Currently leading Enterprise AI Strategy & Architecture at Celanese, where I engineered askchemille.com — a client-facing multi-agent AI platform for polymer grade selection that directly contributed $1.5M in closed sales. Full-stack ownership: NLP-NER models, RAG pipelines, orchestrator/guardrail/validator agents, and CI/CD on Azure.

My edge sits at the intersection of LLM engineering, agentic orchestration, and production MLOps — zero hallucinations in production, $9M+ in measurable business impact, and systems that actually scale.

I hold an ADaSci Certification in Agentic AI System Architecture and an eCornell Data Analytics 360 credential. Open to global opportunities.

Work History

Celanese

Jan 2025 – Present

Chemical / Materials

Sr. AI Engineer / Architect

Engineered askchemille.com for Celanese — a production multi-agent AI platform for polymer grade selection powering the global sales team. Owned end-to-end architecture and delivery.

Contributed to $1.5M closed sales via AI-driven grade recommendation
Improved recommendation accuracy by 26% — GPT-4.1 + LLaMA-3-8B hybrid
Reduced material comparison time by 90% using o1-mini and o3-mini
Cut customer query resolution time by 40%; satisfaction up 15%
Orchestrator, guardrail, validator & assumptions agents in production

LangGraphAzure OpenAIRAG GPT-4.1LLaMA-3NLP-NER Azure DevOpsCI/CD

Acies Global

Apr 2022 – Aug 2024

Enterprise Consulting

Senior Data Scientist

Led GenAI and data science engagements for Google, Waymo, Bose, Under Armour, Kraft Heinz, and Ecolab. Managed a cross-functional team of 4 data scientists.

Identified $3M in Google Ad revenue opportunities through spend analysis
Streamlit app cut analysis time from 2 hrs → 10 min
RAG pipeline reduced query resolution time by 40%
Zero hallucination/jailbreak in production — BERTScore + ROUGE guardrails

GPT-4oRAGBERTScore ROUGEStreamlitBigQuery

Sitel Group

Mar 2021 – Apr 2022

Telecom / UCaaS

Data Scientist

Multilingual helpdesk automation, churn prediction, and intent classification at scale for a Telecom UCaaS client.

60% decrease in MTTR via Deep Translator NLP model (5+ languages)
Generated $136K/year — app licensing + deprecated third-party tools
93% BERT accuracy on 1.5M transcript multi-class classification
29% improvement in retention; 3.5% reduction in churn revenue loss

BERTDeep TranslatorNLPScikit-learn

Wipro

May 2017 – Feb 2021

Enterprise Analytics

Group Lead Analytics

Enterprise analytics — CLV modelling, product recommendation engines, renewal prediction, and BI dashboards for global clients.

RFM model on 2M customers — Gaussian Mixture + Gradient Boosting
85–90% accuracy on renewal prediction models
Saved $224K/year — won PRAGATHI AWARD

Scikit-learnPySparkGradient Boosting Power BITableau

Technical Skills

LLMs & GenAI

GPT-4.1ClaudeLLaMA LangChainLlamaIndexHugging Face RAG PipelinesEmbeddingsLoRA / QLoRA PEFTRLHF

Agentic AI

LangGraphAutoGenCrewAI OpenAI Agent SDKn8nMulti-agent Systems Tool UseMemory ManagementRedis

ML Frameworks

PyTorchTensorFlowScikit-learn TransformersLangFuseFastAPI FlaskMLFlow

Data & Vector DBs

FAISSPineconeWeaviate QdrantBigQuerySnowflake RedshiftSparkKafka

Cloud & MLOps

Azure CloudAzure DevOpsDocker CI/CD PipelinesGitPostmanJIRA

Languages & Viz

PythonSQLJavaScript BashREST / gRPC Power BITableauPlotly

Featured Projects

01 · FEATURED

askchemille.com · Celanese

Celanese Corporation · Jan 2025–Present

Production multi-agent AI platform for polymer grade selection. Built as sole architect — NLP-NER, RAG, orchestrator/guardrail/validator agents deployed on Azure.

GPT-4.1LLaMA-3-8BLangGraph Azure OpenAIRAGNLP-NERAzure DevOps

Sales contribution $1.5M
Recommendation accuracy +26%
Comparison time reduction −90%
Query resolution time −40%
Customer satisfaction +15%
Production hallucinations 0

GenAI Product Intelligence

Acies Global · Google, Bose, Under Armour

GPT-4o review summarisation, RAG support pipeline, BERTScore + ROUGE eval guardrails — zero hallucination in production.

Ad revenue identified $3M
Analysis time (2hrs→10min) −92%
Query resolution −40%

GPT-4oRAGBERTScoreROUGEStreamlit

Multilingual Helpdesk Automation

Sitel Group · Telecom UCaaS Client

Multilingual NLP across 5+ languages. Deprecated third-party tools. 93% BERT classification accuracy on 1.5M transcripts.

MTTR reduction 60%
Annual savings $136K
Retention improvement +29%

BERTDeep TranslatorNLPScikit-learn

Hard Problems Solved

LLM hallucination in production

Implemented BERTScore + ROUGE evaluation pipeline with an automated guardrail agent. Result: zero hallucination or jailbreak incidents in Celanese's production deployment.

Multi-agent orchestration at scale

Designed orchestrator-validator-guardrail pattern using Azure OpenAI Agent SDK. Agents are modular and independently swappable — no cascading failures during updates.

Multilingual helpdesk at scale

Built Deep Translator + custom NLP pipeline across 5+ languages, replacing expensive third-party tools while delivering 60% faster mean time to resolution.

Grade recommendation accuracy