Products

AI projects, ready to run.

Extracted from work I have actually shipped, not written as tutorials. Each one comes with the evaluation harness or measurement tooling that made it trustworthy in the first place.

Templates
Guides

RAG starter kit answering a question with the retrieved source passages shown alongside

Templates

Production RAG Starter Kit

A retrieval pipeline with the evaluation harness already wired in.

A complete retrieval-augmented generation service built the way I build them for clients: structure-aware chunking, hybrid dense and keyword search, cross-encoder reranking, and an evaluation harness so you can prove a change helped instead of hoping it did.

Python
FastAPI
LangChain
pgvector

$149Details

Dashboard showing per-conversation token spend, cache hit rate and model routing decisions

Templates

LLM Cost Control Toolkit

Two cache layers, model routing and hard budget ceilings.

The middleware I wrote after a production LLM feature cost nine times its estimate in week one. Exact and semantic caching, task-based routing between cheap and frontier models, conversation history capping, and per-user token budgets enforced in your own code rather than discovered on an invoice.

Python
FastAPI
Redis
OpenAI

$89Details

Templates

Handwriting OCR Pipeline

Deskew, binarise, segment, recognise, correct.

A transcription pipeline for real-world documents rather than clean scans: automatic deskewing, adaptive binarisation that survives uneven lighting, line segmentation, per-line recognition, and lexicon-based correction with character error rate reporting so you know whether the output is worth reviewing.

Python
OpenCV
PyTorch
Transformers

$119Details

Agent execution trace showing each tool call and decision made while answering a question

Guides

Agent Architecture Blueprint

When to build an agent, and how to keep it bounded.

A written guide and reference implementation covering the decision most teams skip: whether the problem needs a runtime planner at all. Includes a bounded-agent implementation with iteration caps, total tools, full step tracing, and the pipeline alternative side by side so the comparison is concrete.

Python
LangGraph
Pydantic
OpenAI

$59Details

Classification report and confusion matrix used to evaluate a model on imbalanced data

Guides

ML Evaluation Playbook

How to know whether your model is actually any good.

The metrics, splits and sanity checks that separate a model you can deploy from one that merely scores well. Covers imbalanced classification properly, leakage patterns that inflate results, threshold tuning, and how to report findings to people who will make decisions based on them.

Python
scikit-learn
imbalanced-learn
Jupyter

$39Details

TemplatesComing soon

Vision Inference Service

Batched, queued detection serving that holds up under load.

A production serving layer for computer vision models: request batching, a queue that degrades predictably under load, GPU memory management, and model versioning so you can roll a new checkout forward without downtime. Built around YOLO detection but model-agnostic.

Python
PyTorch
FastAPI
YOLO

$129Details

Need something built for your case instead?

These are starting points. If your problem needs a system designed around your data and constraints, that is the consulting work — and it usually starts with a conversation about whether AI is even the right answer.

Next step

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.

Start a project Book a call tasrifnurhimel@gmail.com

Typical reply within 24 hours

Working from Dhaka, Bangladesh · remote worldwide