Hello there, my name is Nikola, and I'm a software developer specializing in backend development and machine learning. With years of experience in both fields, I've developed a strong understanding of the complexities of backend systems and the power of machine learning algorithms.
What I love most about my work is connecting these two worlds. I find it fascinating to explore the endless possibilities that arise from integrating ML into backend systems. It provides an opportunity for research and innovation, while at the same time keeping me grounded in the practicalities of real-life applications.
In addition to my work, I'm always seeking new opportunities to learn and grow. Whether it's taking on new challenges or staying up-to-date with the latest developments in my field, I'm always eager to expand my knowledge and improve my skills.
When I'm not coding, I enjoy spending time with my friends and family, working out, and immersing myself in nature. These activities help me recharge and stay balanced, both in my personal and professional life.
Offering my services as a freelancer with main focus on machine learning (both deep learning and classical machine learning techniques) and backend development with Python. I have good experience with C++ as well.
I have solid experience with Rasa, ElasticSearch, MongoDB, FastAPI and AWS and I'm looking to provide value to my clients. Also, I'm more than eager to learn new things.
- Reveal.ai project
Summary
Reveal.ai is a startup implementing an interview bot. The bot is concerned with gathering data points based on users’ responses to a set of questions.
I was concerned with Rasa chatbot development and the improvement of the given state, both the backend portion and ML portion as well as the whole architecture and how things will interact with each other (on the conceptual level).
Besides the Rasa bot, we’ve had to have a way to present the data to the users so things can be tracked. For this use case, I’ve decided to go with Elasticsearch. I’ve built the backend (AWS Lambda) that was processing MQ events posted by Rasa and storing these in Elasticsearch, SNS, and MongoDB for future processing and retrieval.
Also, there was a need to build super admin functionality for handling the data and I was responsible for building the API. The whole backend was implemented with FastAPI and the infrastructure with Serverless.
- Responsum project
Summary:
Part of a startup named Responsum that is building a new generation of sales helper app. The idea of the app is to integrate multiple different apps under one chatbot. The chatbot integrates with Salesforce, Google and Microsft calendar, Slack and Teams, etc.
On the project, I’ve been tasked with backend as well as ML/NLP.
In order to integrate multiple services, I’ve integrated the Tray.io platform for this use-case as a FastAPI server.
In order to build chatbot functionality I was working with Rasa, sifting through research papers in order to understand the best approaches to solving the problems at hand. I worked on implementing the pipeline for training and testing the process. Also built Elasticsearch queries for some of the object retrievals.