p
princepartho

Partho

@princepartho

Senior Machine Learning Engineer

Bangladesh
Inglese, Bengali, Hindi
Alcune informazioni sono riportate in lingua inglese.
Chi sono
I'm a Senior Machine Learning Engineer with 5 years of production experience building NLP pipelines, LLM-powered chatbots, recommendation systems, and OCR solutions. I've shipped real systems — a restaurant recommendation engine for a food delivery app, an NER-based address extractor for a courier service, and an LLM-driven product tagging pipeline for e-commerce. My work doesn't stop at the model — I deploy using FastAPI and Docker so your solution is ready for production from day one. If you need something that actually works in the real world, let's talk.... Continua a leggere

Competenze

p
princepartho
Partho
offline • 
Tempo di risposta medio: 2 ore

Consulta i miei servizi

Elaborazione del linguaggio naturale
I will build your nlp projects
A partire da100 USD / progettoMaggiori dettagli
visione artificiale
I will extract data from any documents using ocr
A partire da90 USD / progettoMaggiori dettagli

Esperienza lavorativa

Toptal

AI Engineer

Toptal • Full time

Nov 2022 - Present • 3 yrs 9 mos

Developed a personalized food recommendation system from scratch for a food delivery platform using NVIDIA Merlin, candidate generation, ranking workflows, feature engineering, and model deployment. • Built and deployed an ASR-powered travel assistant chatbot integrated into a production mobile application using NLP and conversational AI pipelines. • Designed an LLM-driven e-commerce product tagging pipeline to improve semantic search with validation mechanisms to reduce hallucination and improve tag accuracy. • Implemented image-based product category classification using CLIP embeddings to support visual search and improve product discovery. • Designed and deployed an OCR pipeline for handwritten English documents achieving 86%+ extraction accuracy across 5+ invoice formats • Built an NER-based address extraction system for courier service applications • Improved face verification accuracy through additional preprocessing and validation mechanisms. • Developed scalable ML inference APIs using FastAPI and Docker for production model serving.