AGENTIC AI / GENAI PLATFORM ENGINEER

Hi, I'm Jashwanth

Specializing in governed multi-agent systems, multimodal RAG, and SQL/graph retrieval over enterprise data. Building the future of AI-powered enterprise automation.

% Query Time ↓

% Accuracy ↑

% Faster Proto

Explore My Work ↓ 📄 View Resume

LangGraph

RAG Expert

Multi-Agent

Apps Integrated

Years Experience

Certifications

Publication

Expertise

Technical Skills

🤖

Agentic GenAI Systems

Multi-agent orchestration, task decomposition, MCP servers, multimodal RAG, GraphRAG, guardrails & safety alignment.

Multi-AgentRAGGraphRAGGuardrailsAmazon Bedrock

⚡

Frameworks & Tooling

LangGraph, LangChain, AutoGen, CrewAI, Google ADK, OpenAI Realtime API, Gemini Live API, Hugging Face.

LangGraphLangChainCrewAIAutoGenMCP

🧠

Models & Fine-Tuning

LLMs/VLMs via OpenAI, Anthropic, Gemini, Bedrock. SFT, LoRA/QLoRA, DPO, vision fine-tuning, synthetic data.

LoRAQLoRADPOSFTVision

🔍

Retrieval & Search

Text-to-SQL, text-to-Cypher, hybrid retrieval, ColBERT, reranking, RAPTOR indexing, Pinecone, Qdrant, Elasticsearch.

Text-to-SQLColBERTPineconeElasticsearch

☁️

Cloud & DevOps

AWS (EC2, S3, Bedrock), Azure, Google Vertex AI, Docker, Prometheus, Grafana, LangSmith observability.

AWSAzureVertex AIDockerGrafana

🛡️

Security & Data Engineering

HIPAA/SOC2/GDPR compliance, RBAC, OAuth2/JWT. Databricks, PySpark, Airflow, Snowflake, PostgreSQL.

HIPAADatabricksPySparkAirflowFastAPI

Career Journey

Work Experience

AI-ML Engineer

Aug 2024 – Present

PRI Global

📍 St. Louis, MO, USA

Architected enterprise agentic AI platform (LangGraph) with task decomposition, delegation, and tool-governed execution across 20+ business applications.
Built end-to-end RAG systems combining text-to-SQL, text-to-Cypher, and metadata-driven vector retrieval; improved recall@10 by 15% and reduced latency by 25%.
Designed reusable agent architecture with standardized roles, routing policies, progressive disclosure, and failure recovery.
Implemented retrieval upgrades: semantic chunking, RAPTOR hierarchical indexing, multi-representation embeddings with controlled context compression.
Architected real-time speech-to-speech inbound calling agent with OpenAI Realtime API and Gemini Live API.
Led construction blueprint automation using vision-language agents; improved estimation accuracy by 35%.
Mentored a 3-member AI engineering team, reducing onboarding time by 40%.

Generative AI Engineer

Apr 2024 – Aug 2024

Zero2AI Inc.

📍 New York, USA

Led agentic automation for healthcare statistical programming, converting SAS-to-R workflows and generating ADaM/analysis datasets and TLFs.
Built multi-agent pipelines for SAP/protocol authoring and validation; improved delivery speed by 35%.
Fine-tuned LLMs and vision-language models (SFT, LoRA/QLoRA, DPO); improved dataset transformation accuracy by 22%.
Designed HIPAA-compliant GenAI workflows with secure data handling and RBAC.
Architected and launched AppStorm, a prompt-driven multimodal application framework with sequential/parallel/dynamic agent patterns.
Engineered a prompt-to-app deployment pipeline enabling rapid expansion of enterprise AI capabilities.

Research Assistant - AI

Jul 2022 – Dec 2023

Saint Louis University

📍 St. Louis, MO, USA

Co-authored a peer-reviewed paper on multimodal object detection and voice-based localization in assistive AI (ORAJ, 2025).
Built a real-time object detection pipeline using Grounding DINO/YOLO with BLIP for image-text retrieval and visual QA; improved user efficiency by 20%.
Integrated speech recognition + NLP (spaCy) for voice-activated detection, improving user interaction by 15%.

Data Analyst

Aug 2020 – May 2022

Google (Contract via CSS Corp)

📍 Hyderabad, India

Improved data-driven decision-making by 15% and reduced workload by 30% by automating analytics pipelines with Airflow DAGs.
Performed EDA using Pandas and visualizations (Matplotlib/Seaborn) to identify trends and actionable insights.
Built predictive models using scikit-learn/TensorFlow, improving key metrics by 15%.

AI Intern

Jan 2020 – Jul 2020

Futurenet Technologies Pvt Ltd

📍 Chennai, India

Streamlined Tableau reporting by 30%, reducing turnaround time to 48 hours through automation.
Improved model development cycles, increasing accuracy by 15% and accelerating delivery by 20%.

Academic Background

Education

M.S. in Artificial Intelligence

Dec 2023

Saint Louis University

📍 St. Louis, MO, USA

CGPA: 3.71 / 4.00

Credentials

Certifications

🏅

Databricks Certified Data Engineer Associate

Databricks

🤖

AI Agentic Design Patterns with AutoGen

DeepLearning.AI & Microsoft

🧠

Generative AI with Large Language Models

DeepLearning.AI & AWS

⚡

Big Data with PySpark

DataCamp

📐

Math Skills for Data Science

Duke University

Research

Publications

Leveraging Computer Vision and Natural Language Processing for Object Detection and Localization

Yenugu, J. L. S. et al. (2025). Operations Research and Applications (ORAJ), 12(1).

Developed a voice-driven assistive AI system integrating computer vision, NLP, and speech recognition for real-world object localization.

Computer Vision NLP Object Detection Peer-Reviewed 2025

Document

Resume Preview

📄 Download Full Resume