Tasrif Nur Himel
AI Engineer
Tasrif NurHimel
I design and ship AI systems that survive contact with real users — LLM applications, retrieval pipelines, computer vision and the production plumbing that keeps them running.

Currently
AI EngineerDhaka, Bangladesh
0.0+
Years building AI
0+
Projects delivered
0+
Tools & frameworks
0+
GitHub commits
- Python
- PyTorch
- TensorFlow
- LangChain
- LangGraph
- Hugging Face
- OpenAI
- FAISS
- ChromaDB
- Pinecone
- FastAPI
- YOLOv8
- LoRA
- RAG
- CrewAI
- Streamlit
- AWS
- Docker
- Python
- PyTorch
- TensorFlow
- LangChain
- LangGraph
- Hugging Face
- OpenAI
- FAISS
- ChromaDB
- Pinecone
- FastAPI
- YOLOv8
- LoRA
- RAG
- CrewAI
- Streamlit
- AWS
- Docker
Systems in production, not demos.
A sample of projects running in front of real users — from a bilingual campus assistant to retrieval pipelines and vision models. Scroll over a card, drag, or use the arrow keys to browse.
Agentic AI01DIU Proctor - AI-Assistant Proctor Office Support System
The AI-Assistant Proctor Office Support System is a cutting-edge solution designed to enhance the efficiency of university proctor offices by providing bilingual (Bangla and English) voice and chat-based assistance. This system streamlines communication between students, staff, and the proctor office.AI ModelsVector DatabaseMulti-tenant Architecture
Data & Analytics02Data Analyst Agent
An intelligent, no-code data exploration assistant designed for researchers, data analysts, and educators. This agent allows users to upload CSV or Excel files and ask natural language questions to visualize and understand their data without writing code. It uses LLMs, vector similarity, and prompt-driven workflows to generate charts, summaries, and statistical analyses instantly.LLMsVector SimilarityPrompt-driven Workflows
Generative AI03AI CV Checker: A Smart Recruiter System
A smart HR assistant that automates CV screening, interview question generation, and skill matching using Generative AI. It analyzes CVs against job descriptions, generates relevant questions, and can even auto-email candidates, streamlining the hiring workflow with Flask, Gemini, and LangChain.Generative AIFlaskGemini
RAG04RAG-Based Chatbot Model
A local chatbot system powered by Retrieval-Augmented Generation (RAG), capable of answering questions from both plain and scanned PDF files. Integrates FAISS, OCR, and LLaMA 3.1 (8B) via Ollama for offline LLM inference — making it ideal for data-sensitive or disconnected environments.RAGFAISSOCRWhat I can take off your plate.
Four capabilities that plug into each other — a model is only useful once there is an architecture around it and something serving it.
Work experience.
My journey through AI and machine learning, from exploring data and algorithms to building production-ready AI systems with deep learning, computer vision, GenAI, and LLMs.
- Jan 2026 – Present
AI & Automation Engineer (Full-Time, Augmented)
Banglalink Digital Communications Ltd., Gulshan 1, Dhaka
- Architecting and developing Blink, an advanced RAG-based AI assistant for Banglalink with customised retrieval pipelines, hybrid search, multimodal support, auto skill selection and bilingual (Bangla/English) capability.
- Implemented end-to-end LLM observability and tracing with Langfuse, tracking 100% of AI interactions across all production pipelines and cutting debugging time.
- Built a real-time performance dashboard in Grafana on PostgreSQL for operational insight into the assistant's usage and accuracy.
- Retrieval combines keyword and dense vector search with Reciprocal Rank Fusion (RRF) for more accurate, grounded answers.
- Aug 2025 – Nov 2025
AI Software Engineer (Remote)
Corsult, Valley Stream, NY 11581, United States
- Developed social media automation workflows using n8n/Make.com integrated with OpenAI's GPT-4 API.
- Built an AI-Based real-time Web Call Assistant (OpenAI, WebRTC) who can conversation with users on web calls.
- Built an employee ID card system with Vibe Coding and prompt-engineering
- Worked remotely from Bangladesh, collaborating with US teams and automating operations.
- Jan 2025 – Aug 2025
AI Engineer (Contract)
Daffodil International University, Dhaka
- Designed and deployed AI Proctor, a bilingual (Bangla/English) chat + voice assistant for the university's Proctor Office.
- Built an AI-Based Photo Finder using face recognition and vector similarity search to help users locate their images in large photo albums.
- Led the development of DIU Smart Transport, a digital campus mobility system.
- Ongoing Projects: Currently engaged in R&D on Data Analysis Agent, AI-Powered Photo Finder, and Virtual Try-On Assistant.
- June 2024 – Jan 2025
AI Developer (Project Basis)
Center for Data Science and Research (CDSR), Bangladesh | Remote
- Developed a handwritten OCR system using Microsoft's TROCR model for UNESCO's online education platform targeting Rohingya children.
- Integrated a RAG-based chatbot using Mistral LLM and ChromaDB for document-based question answering.
- Built a YOLOv8-based vehicle number plate detection system with OCR support.
- Worked on multiple LLM-based applications using open-source models (LLaMA, Mistral via Hugging Face).
- February 2022 – Present
AI/ML Journey Started
Self-Learning & Exploration
- Began my journey in machine learning driven by a passion to explore how data and algorithms can solve real-world problems.
- Over the past four-plus years, I've grown into an AI Engineer with hands-on experience in Deep Learning, Computer vision, GenAI, LLMs, Agentic AI and backend development.
- Work spans from building experimental prototypes to leading end-to-end development of production-ready AI systems.
- Always focused on delivering scalable, impactful solutions through intelligent automation and thoughtful design.
Key projects.
The work behind the roles — what each system does, what it is built from, and who it is for.
- Featured Project
LSDS Accelerated Adult Learning (UNESCO)
Python, TROCR, Gradio, Hugging Face
- Developed an OCR model using TROCR for handwritten English input evaluation.
- Enabled automated assessment of children's handwriting on digital platforms.
- Deployed a Gradio-based interface for seamless human-in-the-loop evaluation.
- Fine-tuned and tested on real data from UNESCO's e-learning program.
- Focused on inclusive AI for underserved communities.
- Featured Project
AI Proctor (Daffodil International University)
OpenAI, Langchain, FAISS, FastAPI, Uvicorn, Phidata, Livekit-Agents
- Built a bilingual assistant for university Proctor Offices with both chat and voice support.
- Used Retrieval-Augmented Generation (RAG) to ensure accurate, document-backed answers.
- Developed an admin dashboard for managing student cases, scheduling meetings, and updating policies.
- Integrated Livekit-Agents for real-time voice interactions.
- Empowering universities with 24/7 student support via AI.
- Featured Project
AI CV Checker
Gemini, Langchain, Flask
- Automated CV screening and interview question generation with contextual insights.
- Implemented a skill-matching system to align applicants with job requirements.
- Enabled auto-emailing to HR with candidate summaries and scores.
- Designed a sleek UI using Flask and Bootstrap for recruiters.
- Accelerating recruitment workflows using GenAI.
- Featured Project
Data Analysis Agent
LLaMA 3.1 (Nemo), Streamlit, Python
- Built a natural language interface for exploring Excel/CSV datasets visually.
- Supports custom chart generation (bar, line, histogram, correlation matrix, etc.) via prompts.
- Deployed a Streamlit dashboard for data analysts, researchers, and educators.
- Incorporated error-handling agents to detect and explain user input issues.
- Enabling non-coders to analyze data conversationally.
- Featured Project
AI Photo Finder
dlib, face_recognition, FAISS, Flask, React
- Built a system where users can upload a photo of themselves to find all their images in large photo galleries.
- Used face recognition + vector similarity for high-accuracy matching.
- Powered by FAISS for fast nearest-neighbor image retrieval.
- Developed a React-based frontend with an intuitive image search UI.
- Helping photographers automate large-scale photo delivery for events like convocations.
- Featured Project
Vehicle Number Plate Detection
YOLOv8, EasyOCR, Python, Roboflow
- Developed a system to detect vehicle license plates in images using YOLOv8.
- Used UK-based annotated dataset from Roboflow for training and testing.
- Integrated EasyOCR to extract license numbers with high accuracy.
- Built for potential use cases in campus entry monitoring and traffic logs.
- Combining real-time detection with OCR for smart surveillance solutions.
Competitions won, credentials earned.
- 01
Champion – DIU FSIT Project Showcasing 2024
Smart Transport App – Innovation for Smart City
- 02
Champion – DIU Agentic AI Contest
AI Assistant for Campus Safety & Code of Conduct
- 03
Champion – 4th National Data Science Summit
AI Chatbot Category
- 04
Top 5 Finalist – AI Hackathon by Innovation Conclave BD & BRAC
Competitive AI innovation challenge
- 05
2nd Runner-Up – NASA Space Apps Challenge 2025 (Local Round)
Data Pathways to Healthy Cities and Human Settlements
Certifications
- Building LLM Applications with Prompt EngineeringNVIDIA↗
- Career Essentials in Generative AIMicrosoft↗
- GitHub FoundationsGithub↗
- AWS Academy Graduate - Machine Learning FoundationsAmazon AWS↗
- AWS Academy Graduate - Machine Learning for Natural Language ProcessingAmazon AWS↗
- AI Agent FundamentalsHugging Face↗
- Ethics in the Age of Generative AILinkedIn↗
Have an AI problem worth solving?
Tell me what you are trying to build. I will tell you honestly whether AI is the right tool, what it would take, and how I would approach it.
Typical reply within 24 hours
Working from Dhaka, Bangladesh · remote worldwide