Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m Hema Sri Nakirikanti—a Gen AI developer and ML engineer focused on building multi-modal AI systems and integrating LLMs into full-stack applications. I enjoy designing scalable data pipelines, retrieval-augmented generation, and robust production deployments on cloud platforms, delivering practical AI solutions that solve real-world manufacturing and business problems. I thrive in cross-functional teams, practicing responsible AI with guardrails, evaluation, and user feedback loops to continuously improve model outputs. My work spans research-led model fine-tuning, prompt engineering, and DevOps-driven delivery to ensure reliable, compliant, and measurable results.…Hi, I’m Hema Sri Nakirikanti—a Gen AI developer and ML engineer focused on building multi-modal AI systems and integrating LLMs into full-stack applications. I enjoy designing scalable data pipelines, retrieval-augmented generation, and robust production deployments on cloud platforms, delivering practical AI solutions that solve real-world manufacturing and business problems. I thrive in cross-functional teams, practicing responsible AI with guardrails, evaluation, and user feedback loops to continuously improve model outputs. My work spans research-led model fine-tuning, prompt engineering, and DevOps-driven delivery to ensure reliable, compliant, and measurable results.

HEMA SRI NAKIRIKANTI

AI Developer, AI Engineer, Data Scientist, +3





Hi, I’m Hema Sri Nakirikanti—a Gen AI developer and ML engineer focused on building multi-modal AI systems and integrating LLMs into full-stack applications. I enjoy designing scalable data pipelines, retrieval-augmented generation, and robust production deployments on cloud platforms, delivering practical AI solutions that solve real-world manufacturing and business problems. I thrive in cross-functional teams, practicing responsible AI with guardrails, evaluation, and user feedback loops to continuously improve model outputs. My work spans research-led model fine-tuning, prompt engineering, and DevOps-driven delivery to ensure reliable, compliant, and measurable results.…Hi, I’m Hema Sri Nakirikanti—a Gen AI developer and ML engineer focused on building multi-modal AI systems and integrating LLMs into full-stack applications. I enjoy designing scalable data pipelines, retrieval-augmented generation, and robust production deployments on cloud platforms, delivering practical AI solutions that solve real-world manufacturing and business problems. I thrive in cross-functional teams, practicing responsible AI with guardrails, evaluation, and user feedback loops to continuously improve model outputs. My work spans research-led model fine-tuning, prompt engineering, and DevOps-driven delivery to ensure reliable, compliant, and measurable results.

Available to hire

Hi, I’m Hema Sri Nakirikanti—a Gen AI developer and ML engineer focused on building multi-modal AI systems and integrating LLMs into full-stack applications. I enjoy designing scalable data pipelines, retrieval-augmented generation, and robust production deployments on cloud platforms, delivering practical AI solutions that solve real-world manufacturing and business problems.

I thrive in cross-functional teams, practicing responsible AI with guardrails, evaluation, and user feedback loops to continuously improve model outputs. My work spans research-led model fine-tuning, prompt engineering, and DevOps-driven delivery to ensure reliable, compliant, and measurable results.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

Gen AI Developer at TOYOTA

April 1, 2025 - Present

Developed a Multi-Agentic Reactive (MAR) RAG application using LLMs (Gemini, Claude) with LangGraph to identify manufacturing problems. Built fault-tolerant ingestion pipelines for diverse formats (PDF, images, Excel/CSV, JSON, Markdown, SQL, XDW, video, PPT) with outputs to S3. Leveraged Bedrock-hosted models via SageMaker for high-accuracy extraction and summarization of multimodal content on large EC2 instances. Implemented a hybrid OpenSearch retrieval system with fuzzy text queries and k-NN vector similarity. Created a video preprocessing prototype with OpenCV/FFmpeg for context-aware recommendations. Developed scalable RESTful APIs with FastAPI to integrate LLMs, retrieval logic, and a React frontend, managed credentials with AWS Secrets Manager. Implemented semantic search using OpenAI embeddings and HF all-MiniLM-L6-v2. Containerized the solution with Docker and deployed to Amazon ECS via GitHub Actions CI/CD across dev, staging, and production. Built a RAG quality evaluation w

Gen AI Engineer at Amadeus

February 1, 2025 - October 24, 2025

Developed a Multi AI Agent-based RAG-based Recommendation system using Azure OpenAI GPT-4o to suggest configuration changes, evaluated with RAGAS and DeepEval for generic and code-change scenarios. Employed advanced prompting techniques (one-shot and few-shot Chain of Thoughts) to optimize LLM performance. Designed robust FastAPI REST endpoints to connect LLMs with retrieval mechanisms (BM25Okapi, Azure Search, FAISS) and frontend (Gradio). Used LangGraph for workflow management and state via nodes. Deployed on Azure Web App with Docker and Kubernetes, integrating ACR for image management and GitHub Actions for CI/CD. Implemented Presidio for PII compliance, achieving 60% reduction in data breaches. Implemented unit tests, monitoring, and logging (pytest, Azure Application Insights). Architected scalable data lakes (ADF) and deployed TTS/STT models for voice-based interactions. Built a User Feedback Mechanism for model outputs.

Senior AI Engineer at FLOW

September 1, 2024 - October 24, 2025

Developed a multi-agent RAG-based Recommendation system and configured robust data governance. Managed stateful workflows with LangGraph, implemented data ingestion pipelines, and built a video-based interaction prototype for context-aware recommendations. Deployed on Docker and Kubernetes with CI/CD via GitHub Actions. Implemented PII compliance (Presidio) and built a robust testing/monitoring framework.

Machine Learning Engineer at Texas Tech University

October 1, 2024 - October 24, 2025

Developed a production-ready computer vision solution using transfer learning with VGG16, achieving 95% test accuracy on arts dataset. Automated data extraction with PySpark and Selenium, reducing processing time by 20%. Conducted EDA and built predictive models; performed sentiment analysis and topic modeling. Built websites using HTML/CSS/JavaScript/ASP for faculty and student management; migrated code from C# to ASP.NET. Implemented a reinforcement learning model (DQNs, PPO with RLHF) to detect and mask sensitive tokens.

Machine Learning Engineer at INFOR

December 1, 2022 - October 24, 2025

Collected and cleaned large-scale data using AWS Glue for ETL; handled missing values, outliers, and encoding; stored processed data in S3 and Redshift. Built customer churn prediction models in SageMaker; tuned hyperparameters with GridSearchCV and used K-fold cross-validation. Built distributed ML pipelines with Spark MLlib, Databricks, and TensorFlow on Big Data platforms; optimized SQL queries for data-driven decisions; performance-tuned Spark jobs. Implemented CI/CD for model deployment with Jenkins and AWS Code Pipeline.

Python Developer at HCL Tech

August 1, 2021 - October 24, 2025

Developed and optimized RESTful APIs using Django REST Framework to support e-commerce features; integrated backend services with React frontend for dynamic data rendering and real-time updates. Conducted statistical analysis in Excel to identify patterns; optimized Tableau dashboards for performance; tuned PostgreSQL queries for better response times.