Open to AI/ML Engineering roles

Hello,
Sai here! 👋

I'm an AI/ML Engineer with 3 years of experience designing and deploying intelligent machine learning and Generative AI solutions across customer analytics and enterprise support automation.

I build LLM-powered applications, RAG pipelines, semantic search systems, and Transformer-based NLP models — then ship them to production.

All things AI: Generative AI. RAG Systems. MLOps.
Yep, I do them all.

Find me
Resume reddysaikumar614@gmail.com
Sai Kumar Reddy
3+
Years building
AI/ML systems
AWS
Solutions Architect
Associate
See the systems running →

Systems
in production

Not prototypes. Retrieval platforms and scoring engines that retrieve across technical documentation and rank accounts by risk — containerized, monitored, and retrained on drift.

Example interface mockup: a support assistant returning a context-aware answer with its retrieved source documents ranked by relevance
Example interface — retrieval-augmented technical support
Example dashboard mockup: anonymised accounts ranked by churn risk score with the signal that triggered each
Example interface — churn risk scoring
Four Teledyne business segments feeding the retrieval layer, with FY2025 net sales share for each
Teledyne segment mix — FY2025 net sales
01

Intelligent Technical Support Automation

Teledyne Technologies Inc · AI/ML Engineer · Jul 2025—Present An enterprise RAG platform that retrieves, ranks and generates context-aware answers from technical manuals, SOPs and years of support history — across four engineering domains that share almost no vocabulary.
RAG architecture: knowledge sources through chunking and vector stores, queries through intent classification, retrieval, reranking, agent orchestration and grounded generation
35%
Fewer support tickets
30%
Better search relevance
18%
Higher response accuracy
40%
Faster deployments
View complete case study
02

Customer Retention Intelligence

Colt Technology Services · Junior Machine Learning Engineer · May 2021—Jun 2023 A churn prediction platform over 250K+ customer records, joining usage, billing and support signals into one view — and turning retention from a reaction into a ranked, workable list.
Churn pipeline: source systems through ETL unification, cleaning, feature engineering, model training, precision recall and F1 evaluation, and FastAPI serving
250K+
Records processed
15%
Accuracy improvement
30%
Faster preprocessing
20%
Lower inference latency
View complete case study
03

ragx — Hybrid RAG, Measured

Open source · Python · Personal project Most reference RAG implementations show that a query returns something plausible. This one ships the evaluation: a labelled query set, metrics tested against hand-computed values, and a script that regenerates every published number.
Measured retrieval results: on general prose across 61 queries BM25 scores 0.813 MRR, dense 0.832 and hybrid 0.836 with overlapping confidence intervals; on an identifier-heavy slice of 24 queries BM25 scores 0.889, dense 0.938 and hybrid 0.958
0.836
Hybrid MRR, prose
0.958
Hybrid MRR, identifiers
110
Tests passing
85
Labelled queries
View the code on GitHub
What I build with

The toolkit

From Transformer fine-tuning to cloud-native deployment — the full path from experiment to production.

◆Generative AI & LLMs

LangChainHugging FaceRAGOpenAI APILlamaIndexLangGraphAI Agents

◆ML & Deep Learning

PyTorchTensorFlowScikit-learnKerasTransformersCNNs · RNNs

◆NLP & Vector Search

BERTspaCyPineconeFAISSChromaDBSemantic Search

◆MLOps & Cloud

DockerKubernetesMLflowFastAPIAWS SageMakerCI/CD
Background

Education & certification.

M.S. in Advanced Data Analytics
University of North Texas — Denton, TX
Aug 2023 — May 2025
B.S. in Computer Science
Osmania University — Hyderabad, India
Jun 2018 — May 2021
✦
AWS Certified Solutions Architect — Associate Amazon Web Services

Get in touch

reddysaikumar614@gmail.com
Sai-Kumar-Reddy-Resume.pdf
© Sai Kumar Reddy
AI/ML Engineer · Generative AI · MLOps