|
|
|
|
PROMPT ENGINEER/AI / ML ENGINEER / Data Scientist Resume
|
| Desired Industry: Healthcare |
SpiderID: 84951 |
| Desired Job Location: Fredericksburg, Virginia |
Date Posted: 5/7/2025 |
| Type of Position: Contractor |
Availability Date: 05/07/2025 |
| Desired Wage: |
|
|
U.S. Work Authorization: Yes |
| Job Level: Experienced with over 2 years experience |
Willing to Travel: Yes, More Than 75% |
| Highest Degree Attained: Bachelors |
Willing to Relocate: Yes |
Objective:
SUMMARY
• 9+ years of IT experience with experience in designing, developing, and deploying Machine learning solutions, ETL pipelines, and scalable data processing systems.
• Expertise in developing and deploying AI-driven applications using Generative AI, LLMs, Python, TypeScript, and Java. Skilled in fine-tuning and evaluating LLMs (GPT, Gemini), optimizing ML models, and implementing RAG, Chain-of-Thought Prompting, and Transformer Architecture.
• Proficient in cloud platforms (GCP, AWS) and MLOps, with a focus on scalable AI deployment via data engineering, cloud automation, and CI/CD pipelines. Adept at AI research, machine learning, and cloud-native architectures, delivering robust AI solutions.
• Proven expertise in building predictive models, natural language processing (NLP), deep learning frameworks, and significant data ecosystems.
• Adept at leveraging cloud platforms like AWS, Azure, and GCP for seamless data ingestion, model training, and deployment.
• Strong programming skills in Python, R, and SQL, and a deep understanding of AI/ML algorithms, statistical modeling, and data visualization.
• Passionate about solving complex problems through data-driven insights and cutting-edge technologies.
TECHNICAL SKILLS
Languages & Framework: Java, JavaScript, TypeScript, Python, React.js, Redux, Html5, Css3, Bootstrap, Spring Boot, Node.js, RESTful APIs, Hibernate, MVC
Database: MySQL, PostgreSQL, Microsoft SQL Server, Oracle, MongoDB, Cassandra, Redis
Tools & Technology: PromptFoo, AWS (EC2, S3, Lambda, RDS), Azure (App Services, Functions, SQL Database), Jenkins, GitHub Actions, Azure DevOps, JUnit, Mockito, Selenium, Jest, Enzyme, OAuth2, JWT, Secure Coding Practices, Security Audits, Docker, Kubernetes, Maven
Generative AI & Machine Learning: LLM evaluation, PromptFoo expertise and test suite creation, Prompt optimization, Automated testing, YAML configuration, Model-graded evaluation, Performance metrics, Data driven analysis and deploying LLMs (GPT, BERT, T5, Gemini), Model evaluation techniques using Promptfoo, Retrieval-Augmented Generation (RAG), Chain-of-Thought Prompting (CoT), Transformer Architecture, and Fine-Tuning Methodologies.
Experience:
Elevance health, Dallas TX Sep 2023- Till Date
Prompt Engineer/AI Engineer
Responsibilities
• Designed and optimized AI input prompts, improving response quality by 12%.
• Integrated NLP models into business applications.
• Collaborated with cross-functional teams to enhance AI-driven solutions using ChatGPT and llama.
• Performed code evaluation for multiple LLMs including GPT, Gemini, and BERT, analyzing their achieved responses in regard to performance evaluation alignment.
• CoT prompting as well as reasoning preservation were advanced instruction strategies implemented when crafted and optimized GPT model prompts along with other models.
• Domain-specific task adjustments for hallucination mitigation were incorporated into derived task-metric defined fine-tuning strategies.
• Developed and modeled a dynamic user responsive context AI-powered chatbot/agent by embedding contextual memory loops into dialogue management systems.
• Prompt construction and control of generation relied on attention and transformer architecture awareness to guide their usage for the output.
• -Performance across domains was improved by explored and applied diverse instruction tuning, few-shot learning, CoT prompting, and other frameworks of prompt engineering.
• Factually accurate and reliable outputs were enabled through RAG integration of LLMs with external APIs and vector databases.
• Prompt robustness incrementally improved through edge-case systematic breakdown testing identifying prompt vulnerabilities.
• Utilized embedding retrieval and Retrieval-Augmented Generation (RAG) to optimize multi-turn dialogues as well as response, and the overall relevance of large language models (LLMs) outputs.
• Validated automated prompts through unit tests and A/B testing across heterogeneous models and cycles to enhance prompt performance using defined metrics.
• Increased LLM output reliability and dependability by tuning test driven prompting in PromptFoo which resulted in higher unit test passing rates.
• Created automated testing procedures for prompt verification in the form of assertion and test case granular YAML files within Prompt Foo supervised test cases.
• Conducted structured exploratory testes variations for set output change, compliance with specified parameters, and verbal precision.
• Developed competitive LLM response evaluation standards and goals alongside baseline topic sets and bespoke response-shaping frameworks in PromptFoo to streamline building block formulation.
• Amended LLM unit test outcomes with prompt modifications alongside group evaluation of prompt performance.
• Built and maintained fully automated LLM testing structures with uniform evaluation logic and result reporting using YAML and PromptFoo.
• Initiated model scrutiny using rubrics in PromptFoo which resulted in marked improvements to the models under Prompt Generative Feedback's refine feature's accuracy, precision, and user criterial feedback.
FedEx, Memphis TN Feb 2021- Aug 2023
AI/ML Engineer
Responsibilities
• Automated score generation for assessments using OpenAI GPT models, reducing evaluation time by 50% and ensuring consistent results.
• Developed a dynamic React.js frontend and robust Django backend to support real-time scoring and secure data management.
• Designed and optimized MySQL databases for scalable storage and low-latency access, improving query performance.
• Managed CI/CD pipelines with Jenkins and Docker, streamlining deployment and maintaining system reliability on Azure cloud.
• Built AI-driven NLP pipelines for automated response analysis, fine-tuning models to enhance accuracy and fairness.
• Coordinated sprints and timelines using Jira, ensuring timely feature releases and high code quality.
• Deployed the platform on Azure with auto-scaling policies, ensuring reliability and handling growing user demand effectively.
• Implemented monitoring tools like Prometheus and Grafana for real-time system performance tracking and optimization.
• Designed and implemented an AI-driven recommendation engine to deliver personalized deals and promotions to users based on browsing behavior and historical purchase data.
• Utilized collaborative filtering, content-based filtering, and hybrid recommendation techniques to increase user engagement by 20%.
• Built scalable data pipelines using GCP tools such as BigQuery, Dataflow, and Cloud Storage for real-time data ingestion and processing.
• Conducted exploratory data analysis (EDA) on clickstream and transaction data to identify patterns and trends in user behavior.
• Applied advanced natural language processing (NLP) techniques using spaCy and BERT to extract deal metadata and improve categorization accuracy.
• Integrated the recommendation engine into the mobile app and website through RESTful APIs, ensuring seamless user experiences with real-time response.
• Conducted A/B testing to evaluate the effectiveness of the recommendation system, leading to a 15% uplift in conversion rates.
• Automated model retraining pipelines using Airflow and CI/CD practices, ensuring continuous improvement based on new data streams.
• Designed a real-time dashboard using Looker to monitor key metrics such as recommendation accuracy, click- through rates (CTR), and user engagement.
• Implemented reinforcement learning algorithms to dynamically adjust deal recommendations based on user interaction, maximizing ROI.
• Leveraged customer segmentation techniques to tailor recommendations to user personas, increasing user retention by 25%.
• Collaborated with product managers, software engineers, and marketing teams to align the recommendation system with business objectives.
• Enhanced data quality by integrating advanced data validation steps during preprocessing, minimizing inconsistencies in training datasets.
• Handled peak traffic scenarios (e.g., Black Friday) by optimizing model deployment with auto-scaling on Vertex AI.
• Designed and tested new algorithms for predicting high-demand deals based on location and historical patterns.
• Created synthetic datasets using GANs to augment sparse data during initial recommendation model training phases.
• Incorporated Explainable AI (XAI) tools such as SHAP to ensure transparency in recommendation logic for compliance and stakeholder confidence.
• Presented project updates and insights to executives, showcasing improvements in user engagement and revenue growth through ML integration.
• Conducted root cause analysis for pipeline failures, implementing fault-tolerant solutions using GCP monitoring tools.
Education:
B-Tech – Malla Reddy University, India
Skills:
Languages & Framework: Java, JavaScript, TypeScript, Python, React.js, Redux, Html5, Css3, Bootstrap, Spring Boot, Node.js, RESTful APIs, Hibernate, MVC
Database: MySQL, PostgreSQL, Microsoft SQL Server, Oracle, MongoDB, Cassandra, Redis
Tools & Technology: PromptFoo, AWS (EC2, S3, Lambda, RDS), Azure (App Services, Functions, SQL Database), Jenkins, GitHub Actions, Azure DevOps, JUnit, Mockito, Selenium, Jest, Enzyme, OAuth2, JWT, Secure Coding Practices, Security Audits, Docker, Kubernetes, Maven
Generative AI & Machine Learning: LLM evaluation, PromptFoo expertise and test suite creation, Prompt optimization, Automated testing, YAML configuration, Model-graded evaluation, Performance metrics, Data driven analysis and deploying LLMs (GPT, BERT, T5, Gemini), Model evaluation techniques using Promptfoo, Retrieval-Augmented Generation (RAG), Chain-of-Thought Prompting (CoT), Transformer Architecture, and Fine-Tuning Methodologies.
Candidate Contact Information:
JobSpider.com has chosen not to make contact information available on this page.
Click "Contact Candidate" to send this candidate a response. |
|
|
|
|
|