AI Engineer / LLM & Multi-Agent Systems

I build AI agents that reason, retrieve, and run in production.

ReAct and multi-agent pipelines with retrieval, MCP-grounded citations, and full MLOps — built, evaluated, and deployed. M.S. Data Science & AI, UCM · 4.0 GPA.

See the systems → GitHub Résumé ↓
how my agents think — a real ReAct + multi-agent loop running
queryuser question plannerdecompose · route retrieveFAISS · BGE-large toolsarXiv · BM25 · MCP criticself-check answercited
0.990
MRR@5 retrieval
100%
citation grounded
$0.017
cost per claim adjudicated
5
coordinated agents
featured work

Agentic systems, with the numbers to back them

Every project is laid out like a model card — what it does in plain terms, the metrics that matter, the stack, and where to read the code or hit it live.

↔ drag, swipe, or use the arrows — all 6 projects
FeaturedVLM Agent · Injection-Safe

Multi-Modal Evidence Review — VLM Claim Adjudication

Two-stage VLM agent for insurance claim adjudication, injection-safe by design.

Top 1% · 7/8 injections blockedView details →
FeaturedMulti-agent · MCP

TalentFabric AI — Certification Readiness Platform

5-agent Microsoft Agent Framework workflow with a Critic/Verifier loop and MCP citations.

5 agents · 3 IQ layers · MCP-citedView details →
Agentic · RAG

arXiv Agent — ReAct Research Assistant

ReAct research agent with multi-hop decomposition and a self-critique loop.

MRR@5 0.990 · Precision@5 0.950View details →
Production MLOps

Real-Time Credit Card Fraud Detection

End-to-end MLOps: CI/CD → MLflow registry → Dockerized FastAPI on EC2.

MLflow · FastAPI on EC2View details →
Deep Learning · Multimodal

Multimodal Emotion Recognition — Cross-Modal Transformers

Cross-modal Transformer fusing text, audio, and video for emotion recognition.

60.66% F1 · 3 modalitiesView details →
Time Series · Live

US COVID-19 Mortality Forecasting System

Multi-model mortality forecasting with a live Plotly Dash dashboard.

MAE 138 · live dashboardView details →
stack

What I build with

Grouped by what it does, not alphabetised — so you can read the shape of the work.

GenAI & LLM12 tools
ReAct AgentsMulti-Agent OrchestrationRAGMCPMicrosoft Agent FrameworkCritic/Verifier LoopsFAISSBGEBM25Sentence TransformersGemini APIPrompt Engineering
Deep Learning & NLP12 tools
PyTorchTensorFlowKerasTransformersBERTDistilBERTWav2Vec2ViTHuggingFaceCNNsTF-IDFWord Embeddings
Machine Learning8 tools
Scikit-learnXGBoostRandom ForestLogistic RegressionK-MeansSMOTEFeature EngineeringModel Evaluation
MLOps & Cloud13 tools
MLflowFastAPIDockerGitHub ActionsCI/CDREST APIsAWS EC2S3SageMakerGlueAthenaAzure OpenAIMicrosoft Foundry
Data & Statistics10 tools
PandasNumPySciPyStatsmodelsSARIMAXProphetEDAHypothesis TestingA/B TestingTime Series
Languages & Viz10 tools
PythonSQLRC++JavaPlotlyDashStreamlitMatplotlibSeaborn
about

Systems that are measured, not just demoed

6
systems shipped
Top 1%
16 / 1,773 best
4.0
GPA
255
day DSA streak

I'm an AI Engineer finishing an M.S. in Data Science & AI at the University of Central Missouri (4.0 GPA, Graduate Student Achievement Award). I build LLM and multi-agent systems end to end — retrieval, agent orchestration, evaluation, and the MLOps to ship them.

My through-line is rigor: I don't trust a model I haven't benchmarked. That means citation grounding, eval harnesses, holdout vs cross-validation checks, and reproducible pipelines — the difference between a demo and a system someone can rely on.

Before grad school I shipped NLP and cloud-analytics work across three teams, and I keep my fundamentals sharp through competitive programming. Currently open to AI Engineer and ML roles in the US — available immediately.

roleAI Engineer / ML
degreeM.S. Data Science & AI
gpa4.0 / 4.0
focusAgents · RAG · MLOps
locationUnited States
statusavailable now
experience

Where I’ve worked

Feb 2025 — May 2026
Graduate Laboratory Assistant — Data Science
University of Central Missouri · Warrensburg, MO
  • Designed and delivered 15+ ML lab sessions per semester for 40+ students — Python supervised learning, scikit-learn pipelines, and end-to-end evaluation from ingestion through prediction.
  • Coached students through EDA and visualization workflows with Pandas, NumPy, and Matplotlib while maintaining a 4.0 GPA across full-time graduate coursework and research.
May 2023 — Jul 2023
Data Analytics Virtual Intern
AICTE EduSkills · supported by AWS Academy · Remote
  • Engineered a cloud analytics pipeline — Amazon S3 storage, AWS Glue ETL, and Athena serverless SQL — with automated metadata management across 10K+ semi-structured records.
  • Built predictive models in Python and Amazon SageMaker to surface dataset trends, across a structured 10-week AWS Academy program.
May 2022 — Jun 2022
Data Scientist Intern
Swecha · Hyderabad, India
  • Containerized Telugu speech-recognition pipelines with Docker, eliminating environment-specific runtime failures during production handoff.
  • Fine-tuned speech-to-text models (PyTorch, TensorFlow, NLTK) on dialectally diverse Telugu audio spanning 5+ regional variants for an underserved language community.
education

Academic background

M.S.Aug 2024 — May 2026
Data Science & Artificial Intelligence
University of Central Missouri · Warrensburg, MO
★ GPA 4.0 / 4.0
Coursework
Machine LearningArtificial IntelligenceNeural Networks & Deep LearningNLPStatistical FoundationsAdvanced Data Visualization
B.TechNov 2020 — Apr 2024
Computer Science & Engineering (Hons.)
CMR College of Engineering & Technology · Hyderabad, India
★ GPA 9.06 / 10.0 · First Class with Distinction
Note
Bronze Medal in Academics — ranked top 3 in the B.Tech program.
credentials

Certifications & honors

Certifications

Anthropic
AI Fluency: Framework & Foundations
Jun 2026
Anthropic
Introduction to Model Context Protocol
Apr 2026
Microsoft
Applied Skills — Developing Agents in Microsoft Foundry
Jun 2026
AWS
AWS Academy Graduate — Data Analytics & Cloud Foundations
Jun 2023
Google
Foundations: Data, Data, Everywhere
Feb 2023

Honors & Awards

★
Graduate Student Achievement Award
University of Central Missouri · 2025–2026
★
Finalist — IICC National Coding Competition
Top 1% of 100,000+ participants
★
5★ HackerRank · 4★ CodeChef
Data Structures & Algorithms
★
GeeksforGeeks POTD — 255-day streak
Consecutive days solving algorithmic problems
available immediately · open to AI Engineer / ML roles · US
Let’s build something that ships.

If you’re hiring for AI Engineering, ML, or Applied AI and want systems that are evaluated and deployed — not just demoed — let’s talk.

Email me Résumé ↓ LinkedIn GitHub Medium