Daniel is available for hire

Daniel Tunnermann

Verified Expert in Engineering

Back-end Developer

Location

Anápolis - Goiás, Brazil

Toptal Member Since

October 5, 2022

Daniel is a software engineer with over a decade of experience delivering back-end and full-stack projects. With a master's degree in computer science, he incorporated machine learning into his projects, including natural language processing and speech models. Daniel's primary programming languages are Python and Java, but he is adept at applying other skills to deliver results.

Back-end Software Engineering Algorithms Natural Language Processing (NLP)Chatbots OpenAI GPT-4 API Python FastAPI Git Programming Artificial Intelligence (AI)Machine Learning Linux SQL Databases Retrieval-Augmented Generation (RAG)

Portfolio

Akad Seguros

Python, Chatbots, Artificial Intelligence (AI), Databases, Machine Learning...

Department of Federal Revenue of Brazil

Java, Python, Artificial Intelligence (AI), Machine Learning, FastAPI, SQL...

CyberLabs

Python, Neural Networks, Artificial Intelligence (AI), PyTorch, APIs, Git...

Experience

Algorithms - 12 years Data Structures - 12 years Python - 6 years Natural Language Processing (NLP) - 4 years Machine Learning - 4 years Chatbots - 4 years OpenAI GPT-4 API - 1 year Retrieval-augmented Generation (RAG) - 1 year

Availability

Part-time

Preferred Environment

Visual Studio Code (VS Code), Linux, Python 3, FastAPI, SQL, PyTorch, NumPy, Pandas, Scikit-learn

The most amazing...

...application I've developed is an intelligent chatbot with more than a million users.

Work Experience

Senior AI Engineer/Software Engineer

2022 - PRESENT

Akad Seguros

Developed a chatbot with OpenAI GPT LLMs (WhatsApp with a Python back end) for insurance brokers that is serving thousands of users per month. The chatbot uses Retrieval Augmented Generation (RAG) to answer with info from internal docs.
Developed an AI tool for human customer services transferred from the chatbot, including a front end in JavaScript and a back end in Python. It connects to Meta WhatsApp API to continue conversations where the chatbot handles control to the human.
Developed a question-answer tool with OpenAI GPT LLMs (a Python back end) for internal users to answer customer service calls. The tool uses Retrieval Augmented Generation (RAG) to reply with info from internal docs, including previous customer interactions.
Developed autonomous agents with OpenAI LLMs and LangChain to process previous customer services, summarizing and extracting information, checking correctness and attendance to guidelines, and feeding the internal Q&A tool.

Technologies: Python, Chatbots, Artificial Intelligence (AI), Databases, Machine Learning, Natural Language Processing (NLP), OpenAI, Large Language Models (LLMs), Retrieval-augmented Generation (RAG), LangChain, APIs, Git, Back-end, OpenAI GPT-4 API, OpenAI GPT-3 API, System Design, Architecture

Senior Software Engineer

2006 - PRESENT

Department of Federal Revenue of Brazil

Developed question-answer chatbots used by millions, reducing the need for human experts, which is particularly helpful in COVID-19 times, using Python, PyTorch, Hugging Face, and FastAPI.
Built a system in Java to control international collaboration demands that helped keep track of deadlines and the history of the institution's global partnerships.
Engineered a predictive model for fraud detection in tax returns using Python, Project Jupyter, and scikit-learn.
Worked as the institution's chief of development, implementing a software development process, standardizing tools, and managing development projects.
Taught data science and AI courses for dozens of employees. During the pandemic, these courses were held online.

Technologies: Java, Python, Artificial Intelligence (AI), Machine Learning, FastAPI, SQL, APIs, Git, Back-end, OpenAI GPT-4 API, OpenAI GPT-3 API, System Design, Architecture

Senior Software Engineer - Machine Learning

2021 - 2022

CyberLabs

Built neural text-to-speech models using Python and PyTorch, with a mean opinion score of 4.4 out of five, comparable to human speech.
Created a speaker verification model with 99.6% accuracy in the internal dataset. Used Python and PyTorch for training the model and TorchServe to serve it.
Developed the back end using Python to serve these models with APIs hosted on AWS.
Designed the preprocessing pipeline for the speaker verification model, with deep fake detection with 96% accuracy, noise reduction, and audio length validation.

Technologies: Python, Neural Networks, Artificial Intelligence (AI), PyTorch, APIs, Git, Back-end, System Design, Architecture

Experience

Chatbot for the Brazilian Tax Agency

http://www.rfb.gov.br

Chatbot that answers questions about Brazilian federal taxes and customs. I developed the AI, the back end to serve the models, contributed to the back end of the bot itself, helped develop a compiler for a language used in bot specifications, and helped set up the infrastructure to run the product.

Skills

Languages

Python, SQL, Java

Other

Software Engineering, Algorithms, Data Structures, Natural Language Processing (NLP), OpenAI GPT-4 API, Chatbots, APIs, Back-end, OpenAI GPT-3 API, FastAPI, Retrieval-augmented Generation (RAG), System Design, Architecture, Programming, Artificial Intelligence (AI), Neural Networks, Machine Learning, Speech Synthesis, GPT, Generative Pre-trained Transformers (GPT), OpenAI, Large Language Models (LLMs), LangChain

Tools

Git, Docker Swarm

Frameworks

Spring

Libraries/APIs

PyTorch, NumPy, Pandas, Scikit-learn

Platforms

Visual Studio Code (VS Code), Linux, Docker

Storage

Databases, MongoDB

Education

2019 - 2021

Master's Degree in Computer Science

Federal University of Goiás - Goiânia, Goiás, Brazil

2001 - 2005

Bachelor's Degree in Computer Science

University of Brasília - Brasília, Federal District, Brazil

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring