class TishaChoksi:
role = "AI/ML Engineer | Data Scientist"
location = "Surat, India 🇮🇳"
education = "B.Tech in AI & Data Science — SCET (9.375 CGPA)"
currently = "AI/ML Developer @ Keeva Diamonds"
focus = [
"Production RAG Pipelines & LLM Orchestration",
"Agentic AI Systems (LangChain / LangGraph)",
"Hallucination Detection & AI Reliability",
"Computer Vision (OpenCV / YOLO / MediaPipe)",
"FastAPI Backend for AI Inference",
]
ask_me_about = ["GenAI", "RAG", "LangChain", "FastAPI", "Fine-tuning", "NLP"]
fun_fact = "I built a GAN suite AND a blockchain escrow AI in the same semester 🚀"
|
AI middleware for monitoring & tracing RAG pipelines across retrieval, prompt orchestration, and LLM response flows. Reduces hallucination risks in enterprise AI workflows. Stack: |
Multi-stage verification pipeline using semantic similarity scoring, claim extraction, and context-grounded validation to assess LLM response accuracy. Stack: |
|
AI-driven voice automation agent for lead engagement. Integrated STT + NLP pipelines for real-time conversational experiences with CRM backend. Stack: |
Transformer-based NLP for blockchain escrow dispute resolution. Processed 1000+ transactions, reduced resolution time by 40%. Stack: |
|
Conditional Perceptual Adversarial VAE for realistic facial age progression & rejuvenation. Integrated XAI: Saliency maps, LIME, DeepSHAP. Stack: |
Real-time computer vision pipeline for vehicle speed estimation from highway footage using object tracking with annotated video output. Stack: |
💡 I write about GenAI, RAG architectures, LLM reliability, and applied AI engineering
