Avleen Mehal

AI Researcher

Raleigh, North Carolina, United States3 yrs 3 mos experience

AI ML PractitionerAI Enabled

Key Highlights

Expert in end-to-end AI system design and deployment.
Proven track record in enhancing user experience and engagement.
Strong background in predictive analytics and real-time data processing.

Stackforce AI infers this person is a SaaS-focused AI/ML engineer with expertise in backend development and data analytics.

Contact

avleen.mehal@gigamon.com LinkedIn

Skills

Core Skills

Generative AiLarge Language Models (llm)Retrieval-augmented Generation (rag)Natural Language Processing (nlp)Data ScienceSoftware DevelopmentWeb Development

Other Skills

AWS SQSAmazon Web Services (AWS)Apache KafkaAthenaAtlasAutonomic ComputingC++Cascading Style Sheets (CSS)Collaborative FilteringComputer ScienceCufflinksCustomer AnalysisData AnalysisData IngestionData Mining

About

I design and deploy end-to-end AI systems across LLMs, RAG pipelines, and predictive analytics, combining strong ML fundamentals with real-world software engineering at scale. My work spans research-driven modeling to enterprise-grade deployment on cloud-native infrastructure. I’m particularly strong in LLM orchestration, retrieval-augmented generation, prompt optimization, and ML system design, with hands-on experience using OpenAI, HuggingFace, LlamaIndex, LangChain, PyTorch, TensorFlow, and AWS. Previously, I built large-scale backend and data platforms at Expedia Group, translating ML insights into measurable business impact.

Experience

Gigamon

2 roles

AI/ML Engineer

Sep 2025 – Present · 6 mos

Intern

May 2025 – Aug 2025 · 3 mos

Developed a unified generation framework integrating OpenAI, Ollama, and vLLM providers, featuring tool-based interactions via the MCP and intelligent query enhancement, improving contextual accuracy by 20%.
Designed a dynamic document RAG ingestion service using LlamaIndex to ingest data in real-time, converting complex technical documents into structured LLM-readable Markdown, reducing ingestion duration by 60%.
Evaluated copilot responses using the DeepEval framework and performed prompt tuning with the dspy framework, improving answer relevancy and context recall scores to 94% and 92%.

OpenAIOllamavLLMMCPLlamaIndexGenerative AI+1

Nc state data science and ai academy

Data science consultant | Graduate Research Assistant

Feb 2025 – May 2025 · 3 mos · Raleigh, North Carolina, United States

Providing Data Science consultancy services to clinical researchers associated with NC TraCS institute
Working with clinical data analytics tools like OMOP, Atlas and Athena on EHR like data
Applying data mining methods, statistical analysis and machine learning techniques for clinical research

OMOPAtlasAthenaData Science

Expedia group

3 roles

Software Development Engineer II

Oct 2022 – Jul 2024 · 1 yr 9 mos

Led a project enabling a client-facing capability with GraphQL and Elasticsearch to retrieve aggregated review scores, enhancing the experience for over 36% of property managers with around 800k listed properties on Expedia.
Part of Direct Connect initiative, building multiple microservices in Maven exposing REST apis to fetch and store data into the inventory, further enabling Cache Pricing near real-time for faster loading improving user experience.
Developed a retry mechanism for the services utilizing AWS SQS and Lambdas to create more efficiency.
Increased adjusted PSR (i.e. Purchase Success Rate excluding genuine errors like payment failure, inventory error, booking errors) from 86% to 95%. Reduced cabin availability error rates by at least 65%.

GraphQLElasticsearchMavenREST APIsAWS SQSLambdas+1

Back End Developer

Promoted

Jul 2021 – Jul 2024 · 3 yrs

Architected a predictive analytics platform integrating GraphQL and Elasticsearch, enabling real-time demand forecasting insights that improved client-customer engagement by 28%.
Designed end-to-end REST APIs with Java and Spring in an agile environment for real-time pricing, centralizing inventory data across 800k hotel properties to generate $185M incremental gross booking value.
Implemented scheduling jobs in Jenkins, orchestrating automated ETL workflows to integrate real-time lodging inventory from multiple API endpoints into AWS S3, processing 2M+ records daily, updating ~50% of stale data.
Developed a real-time data transfer publisher-consumer system in an Agile framework using Apache Kafka, automating data ingestion and preprocessing for 200K+ daily events, which augmented data accessibility by 60%.
Automated CI/CD workflows for model retraining, validating, and A/B testing, accelerating deployment cycles by 40%.

GraphQLElasticsearchJavaSpringJenkinsApache Kafka+1

Software Development Engineer

Jul 2021 – Oct 2022 · 1 yr 3 mos

Developed an automated data pipeline using Gradle and Apache Kafka for streaming and processing client data in XSD, XML, and JSON formats.
Implemented a scheduler and event listener in a grade run application, enabling timely data reading, eliminating manual efforts, and enabling easy onboarding for consumers.

GradleApache KafkaSoftware Development

Career launcher

Development Intern

May 2020 – Jul 2020 · 2 mos

Developed application forms for new courses, including both front-end and back-end components. Utilized MongoDB and Node.js for backend development and React for the front end, creating a seamless and intuitive user experience

MongoDBNode.jsReactWeb Development

Engifest dtu

Co-Head

Dec 2018 – Feb 2019 · 2 mos · Delhi, India

I worked for the Digital Marketing Team, where we used to get media partners for our fest and also publicize the fest to gather the footfall of 40k+, and managed and led a team of 30 other members.