Abhinav Reddy
Ai Ml Engineer @Equinix
Signup · Get unlimited contacts
WORK HISTORY
Ai Ml Engineer @Equinix
Redwood City, CA, US
Designed and deployed an Incident Report Summarization & Validation platform using RAG + Azure OpenAI (GPT-4), achieving 95% accuracy in field extraction and reducing manual triage time by 60%Developed an agent-based architecture with ReAct-style Extraction and Validation agents in a RoundRobinChat loop, increasing data completeness by 40% and reducing output errors by 35%Built a vector search knowledge base in FAISS/Azure Cognitive Search, indexing 10K+ historical incidents and runbooks, improving LLM contextual accuracy by 28% and halving hallucination ratesIntegrated with Azure Monitor, AWS CloudWatch, and CMDB APIs using secure SAML-based AWS authentication, cutting SLA validation and outageverification time from 30 minutes to under 2 min
EDUCATION
University of Bridgeport
Master of Science - MS
CMR Institute of Technology, Hyderabad
Bachelor of Technology - BTech
ABOUT ABHINAV REDDY
AI/ML Engineer with 4+ years of experience delivering production-grade machine learning solutions across cloud, edge, and enterprise platforms. Skilled in building agentic AI systems, fine-tuning LLMs (e.g, LLaMA-2), and deploying Retrieval-Augmented Generation (RAG) pipelines for document automation and intelligence. Known for reducing latency, improving accuracy, and driving efficiency through intelligent system design. Currently focused on developing multi-agent LLM workflows, scalable MLOps pipelines, and GenAI solutions using LangChain, FastAPI, Docker, and Triton Inference Server. Key Skills - • LLMs & GenAI: LangChain, Transformers, RAG, LLaMA-2, LoRA, QLoRA, Prompt Engineering • NLP: Named Entity Recognition, Summarization, Document Classification, Vector Databases (FAISS, Pinecone, Milvus) • MLOps & Deployment: Docker, Kubernetes, FastAPI, MLflow, Ray Serve, Triton, Airflow • Computer Vision: YOLOv5, ResNet, CLIPSeg, TensorRT, OpenCV • Cloud & Data: Azure, AWS, Redis, PostgreSQL, Snowflake, Databricks Always open to collaboration and meaningful conversations around AI, automation, and real-world deployment. Let’s connect!
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.