26.9124° N · 75.7873° E · Jaipur, India|ONLINE

TANISHA
YADAV

AI/ML Engineer · GenAI · Backend · RAG Systems

Building intelligent systems that bridge cutting-edge AI research and production-ready applications. Specialising in Retrieval-Augmented Generation, semantic search, and scalable backend APIs.

50K+
Doc Chunks Processed
89%
Answer Relevance
18%
Hallucination Reduction
<120ms
API Latency
RAG Pipelines LLaMA 3 FastAPI · REST APIs FAISS · Pinecone · Chroma Python · C++ · Java Docker · Kubernetes · AWS Anomaly Detection Semantic Search PostgreSQL · MySQL LangChain · HuggingFace RAG Pipelines LLaMA 3 FastAPI · REST APIs FAISS · Pinecone · Chroma Python · C++ · Java Docker · Kubernetes · AWS Anomaly Detection Semantic Search PostgreSQL · MySQL LangChain · HuggingFace
01 — Core Systems

Technical Arsenal

🤖
Generative AI
RAG PipelinesLLaMA 3Sentence-BERTPrompt EngineeringSemantic SearchLangChainHuggingFaceEmbedding Pipelines
🧠
ML / Deep Learning
PyTorchScikit-learnLSTMsAutoencodersIsolation ForestLightGBMNLPFeature Engineering
⚙️
Backend & APIs
FastAPIREST APIsStreamlitPythonC++JavaBashSolidity
🛸
Cloud & DevOps
DockerKubernetesAWS EC2/S3LinuxGit · GitHub
🔭
Vector Databases
FAISSPineconeChromaCosine SimilarityVector Retrieval
🗄️
Databases
MySQLPostgreSQLSQLDBMS
02 — Mission Log

Where I've Served

JUN 2025 — AUG 2025
AmpleLogic
AI / ML Intern
  • Built transformer-based retrieval pipelines processing 50K+ document chunks for enterprise knowledge systems.
  • Improved semantic search relevance by 10% through optimised embedding-based retrieval workflows.
  • Reduced LLM hallucination by 18% using retrieval filtering and context grounding techniques.
  • Developed scalable document processing pipelines for AI-driven knowledge extraction.
  • Deployed FastAPI inference APIs achieving sub-120ms response latency.
  • Containerised ML services using Docker enabling portable, reproducible deployments.
03 — Deployments

Things I've Launched

// Featured Mission

DocuMind AI

End-to-end document intelligence system. Extracts text from PDFs, generates context-aware questions via NLP pipelines, detects document similarity through embedding-based comparison, and enables scalable Q&A — all over a FastAPI backend.

LLaMA 3Sentence-BERTFAISSFastAPIRAGNLP
Mission Details
89%
Answer Relevance Score
RAG
Scalable Architecture
18%↓
Hallucination Rate
01
02
// Agent Automation

Clara Agent

AI-powered automation agent that processes natural language requests and executes workflow tasks. Modular backend with task scheduling, productivity automation, and integration with scheduling systems.

LLM AgentNLPWorkflow OrchestrationFastAPI
GitHub
03
// Industrial ML

AnaVerse 2.0

Industrial anomaly detection using ensemble Isolation Forest + Autoencoders. 40+ temporal features engineered from sensor data, with sliding-window inference for real-time monitoring.

Isolation ForestAutoencodersPyTorchTime-Series
GitHub
04
// Cybersecurity AI

Network Intrusion Prevention

Deep MLP classifiers trained on large-scale network traffic datasets. Identifies malicious intrusion patterns with 91% accuracy. Feature extraction and preprocessing pipelines for cybersecurity analysis.

Deep MLPPyTorchNetwork SecurityFeature Extraction
GitHub
04 — Core CS

Problem Solving

Algorithms & Data Structures

Strong command of graphs, dynamic programming, greedy algorithms, and algorithm optimisation. Experienced in improving efficiency through time and space complexity analysis.

System Design

Grounded in OOP, operating systems, DBMS principles, and system design. Builds scalable, maintainable architectures with clean separation of concerns.

Backend Engineering

Designs and implements REST APIs with FastAPI, handles asynchronous workflows, and integrates relational databases (MySQL, PostgreSQL) in production pipelines.

Relevant Coursework

Data Structures · Design & Analysis of Algorithms · DBMS · Operating Systems · Machine Learning · Deep Learning · Computer Networks

RAG_SYSTEM.py
vector_db.init()
llama3.generate()
deploy.docker()
05 — Profile

Who I
Am

Computer Science undergraduate specialising in AI systems, backend development, and document intelligence pipelines. I bridge the gap between research-level GenAI and production-grade software — from enterprise RAG systems to real-time anomaly detectors.

Currently pursuing B.Tech at Manipal University Jaipur (2022–2026), with hands-on experience deploying ML services at scale using FastAPI, Docker, and AWS.

Manipal University Jaipur
B.Tech — Computer & Communication Engineering
2022 — 2026