You will get a cost-effective NLP solution optimized for your domain — no LLM needed
Project details
Most NLP tasks don't need expensive LLMs. I build fast, lightweight models (BERT, XLM-R, XGBoost) fine-tuned specifically for your domain and language — delivering real-time inference at a fraction of LLM costs. With 10+ years of experience across different languages and domains, I specialize in making NLP work where off-the-shelf tools fail: less-known languages, domain-specific data, and production environments processing millions of documents. You get a model that's accurate, fast, affordable, and truly yours.
Machine Learning Tools
BERT, fastText, MLflow, NLTK, NumPy, pandas, Python, Python Scikit-Learn, PyTorch, scikit-learn, SciPy, Stanford CoreNLP, TextBlob, Vertex AI, Word2vec, XGBoostWhat's included
| Service Tiers |
Starter
$1,500
|
Standard
$3,500
|
Advanced
$7,000
|
|---|---|---|---|
| Delivery Time | 7 days | 14 days | 21 days |
Number of Revisions | 2 | 3 | 5 |
Number of Model Variations | 2 | 4 | 6 |
Number of Scenarios | 2 | 4 | 6 |
Number of Graphs/Charts | 3 | 8 | 15 |
Model Validation/Testing | |||
Model Documentation | - | ||
Data Source Connectivity | - | - | |
Source Code | - |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$400 - $3,500
Additional Revision
+$150
Additional Model Variation
(+ 3 Days)
+$300
Additional Scenario
(+ 2 Days)
+$250
Additional Graph/Chart
(+ 1 Day)
+$50
Model Documentation
(+ 2 Days)
+$200
Data Source Connectivity
(+ 2 Days)
+$600
Source Code
(+ 1 Day)
+$500About Dicle
Senior NLP, ML Specialist | Generative AI Engineer | Data Scientist
Istanbul, Turkey - 10:51 pm local time
My specialty is bringing NLP to life for different languages, especially low-resource ones. It's a space I'm deeply passionate about, and one where I've seen firsthand how much of a difference thoughtful, language-aware engineering makes.
My primary fields:
→ LLM Evaluation & Fine-tuning
→ Multilingual RAG Pipelines
→ BERT & Transformer Models
→ End-to-End NLP Consulting
A little about me: I have an MSc degree in Computer Science from Boğaziçi University and PhD-level coursework in NLP and computational linguistics. I run Lucitext, my NLP consulting practice, and I speak four languages — which gives me not just technical knowledge but real intuition for how languages work.
I do my best work on projects where language complexity is the core challenge, not an afterthought. If your data is multilingual, your users speak underserved languages, or your models need to go beyond English — I'd love to hear about your project.
Steps for completing your project
After purchasing the project, send requirements so Dicle can start the project.
Delivery time starts when Dicle receives requirements from you.
Dicle works on your project following the steps below.
Revisions may occur after the delivery date.
Discussion
Problem definition by the client is elaborated on together and the problem is redefined in technical and AI-based terms by me. Steps, time requirements, technical details and communication options settled.
Development
The problem solution is structured, the system is designed and technical implementations (dataset collection, model development, experiments) are done. Documentation is also ready. The client is informed.