Shinde Aditya
Results-oriented Machine Learning Enthusiast | AI Engineer | GenAI Systems & Inference Optimization
Hyderabad
1+ year of work experience
Aug 20259 mosApr 2026
Architected Multimodal Streaming Pipelines: Engineered an interactive, real-time AI Avatar system utilizing state-of-the-art Speech-to-Video (S2V) synthesis frameworks and streaming LLM outputs; achieved sub-250ms audio-visual synchronization latency for production educational platforms.
Optimized Advanced RAG Implementations: Built and deployed an enterprise-grade RAG pipeline using LlamaIndex and Pinecone; implemented hybrid search (sparse/dense), re-ranking models (Cohere), and metadata filtering, reducing hallucination rates by 35% and improving context retrieval accuracy by 42%.
Designed Agentic Workflows: Developed an asynchronous, multi-agent AI system using LangChain to automate complex data analysis tasks; integrated tool-calling loops and self-correction mechanics, increasing autonomous task completion rates to 94%.
Engineered High-Throughput API Serving Layers: Built a containerized, asynchronous FastAPI serving layer utilizing Docker and AWS EC2 to handle high-volume concurrent LLM inference requests; optimized memory allocation and request batching, improving system throughput by 55%.
Automated MLOps & Evaluation Pipelines: Established automated CI/CD deployment tracks via GitHub Actions paired with Weights & Biases for monitoring model drift and response latency, cutting production deployment time from days to minutes.
Feb 20245 mosJun 2024
Built Automated Data & Ingestion Pipelines: Engineered high-performance ETL data pipelines using Python and SQL to extract, clean, and vectorize massive unstructured datasets, reducing downstream embedding generation latency by 45%.
Deployed Predictive Vision Models: Configured and validated real-time object detection models using YOLO and OpenCV for spatial monitoring; optimized model edge-inference layers to maximize frame-rate stability across consumer-grade hardware.
Fine-Tuned Specialized Embeddings: Collaborated with senior engineering teams to fine-tune domain-specific sentence embeddings; executed comprehensive hyperparameter optimization routines that scaled semantic search relevance across internal databases.
Shinde’s AI Stack
Education
Bachelor of Technology, Data Science
Woxsen University
20214 yrs2025
12, Mathematics and Computer Science
Urbane Junior College
20192 yrs2021
Join Shinde and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal Boss