Dive deeper. Aim higher.
At Abysalto, that’s not just a motto — it’s how we work. We build serious tech for a variety of clients, but we keep things simple, fast, and focused. We’re a team driven by determination, expertise, and courage — and we’re looking for someone who shares that mindset. Someone ready to take ownership, solve real challenges, and make an impact where it matters. Ready to dive in? Join us as a AI/ML Engineer!
🔷What you will do?
- Research, train, finetune, prototype, and production-ise AI models; from LLMs, custom NLP and classical ML models (e.g., LSTM, XGBoost, Random Forest), to vision and multimodal models (e.g., OpenCV, object detection, YOLO, VLMs)
- Build and maintain Retrieval-Augmented Generation (RAG) pipelines using vector databases (FAISS, Pinecone, Milvus, etc.)
- Fine-tune and optimize models with PEFT/LoRA, quantization, and efficient inference engines (vLLM, llama.cpp)
- Evaluate model performance with custom benchmarks; iterate based on business KPIs and user feedback
- Package and deploy models through Docker/Kubernetes and CI/CD pipelines, working closely with DevOps and backend developers
- Collaborate with backend, front-end, and product teams to translate real-world problems into AI solutions
- Document your approach, share learnings, and champion best practices in MLOps and prompt engineering
- Contribute ideas on architecture, MLOps best practices, and new AI capabilities that can delight our customers
- Take ownership of your work, from concept to deployment — and refine it based on real-world feedback
🔷What we expect from you?
- 1-3+ years of hands-on Python for ML/AI (NumPy, pandas, scikit-learn, etc.)
- Production experience with PyTorch or TensorFlow and the Hugging Face Transformers ecosystem
- Familiarity with modern LLM inference tooling (vLLM, llama.cpp, TGI, Triton) and GPU workflows
- Demonstrated success building RAG systems end-to-end (data ingestion, embedding generation, retrieval, prompting)
- Practical know-how of PEFT/LoRA fine-tuning and training of transformers, BERT-style or embedding models
- Solid understanding of NLP tasks: NER, summarization, semantic search, vector similarity
- Comfortable with Git, Docker, and at least one cloud platform (AWS, Azure, or GCP)
- Clear, concise communication in English and a collaborative, curiosity-driven mindset
🔶Nice-to-haves
- Computer-vision hands-on exp (OpenCV, OCR, object detection/segmentation)
- Deep optimization techniques: quantization, KV-cache reuse, GPU offloading
- Experience designing or consuming Knowledge Graphs (Neo4j, RDF/SPARQL)
- Prompt-engineering tricks for LLMs (system vs. user roles, chain-of-thought, few-shot, etc.)
- Familiarity with MLflow, Weights & Biases, or similar MLOps platforms
*We’re hiring at junior to mid-level. If you’re at the start of your career but have strong personal/academic projects and eagerness to learn, we’d still like to hear from you!
🔷What we offer?
- Work in an experienced and encouraging team
- Continuous professional development through training and conferences
- Flexible working hours with the option of hybrid work
- Work in an agile environment following SCRUM methodology
- Pleasant and relaxed work environment with various perks (top-quality Herman Miller Aeron chairs, high-end equipment, discounts with partner companies)
- All perks and benefits can be found on our career page
We solve complex technological challenges in order to simplify and improve everyday lives of millions of people. Our goal is to become a leader in the software industry, recognized for excellence and quality.
If you're ready to shape what’s yet to be — send us your CV and a short note about or a GitHub link to a Python or ML project you’re proud of.
Apply via the link below.
We look forward to meeting you!