Machine Learning Engineer - Computer Vision & Model Serving

About the role

VOSKER, a provider of surveillance solutions for remote-area monitoring, is recruiting a Machine Learning Engineer for its Reconeyez solutions. The role owns and evolves the models and ML infrastructure behind the actor-detection and visual-verification pipeline: the object-detection models that flag intrusions, the duplicate-suppression logic that prevents false alarms, and the next generation of vision-language models for richer scene understanding. It is a hands-on engineering role focused on training, optimizing and deploying models in production.

Key responsibilities

• Train, fine-tune and evaluate computer-vision models (object detection, image quality, static-object/duplicate suppression) on real camera imagery.
• Own the model-serving pipeline: package models into NVIDIA Triton ensembles (DALI preprocessing, TensorRT inference, post-processing), build and deploy TensorRT engines, manage the model repository and no-downtime reloads.
• Build and curate datasets through ingestion, labelling and quality control using FiftyOne (Voxel51) and Label Studio, fixing the data issues that affect accuracy.
• Develop evaluation harnesses such as regression suites, A/B comparisons and metrics tied to real detection quality.
• Extend the self-hosted VLM/LLM stack (vLLM and similar), build retrieval- and tool-using agents, and integrate them into engineering and product workflows.

Required profile

• Mid-level engineer (2-5+ years of relevant ML engineering experience).
• Strong Python and the modern ML stack, including PyTorch, model training and fine-tuning, and notebook-driven experimentation.
• Practical computer vision experience and experience taking models to production, not just training them.
• Self-starter who can solve ambiguous accuracy problems independently, and who cares about rigorous evaluation and edge cases.

Required skills

• Nice to have: NVIDIA Triton, TensorRT, DALI, FiftyOne, Label Studio, vLLM, LoRA, RAG, CometML/Opik, NATS/gRPC.
• Nice to have: edge/IoT inference, anomaly detection, or agent-building experience.

What we offer

• Fast-growing business and a fantastic office in Tallinn.
• Down-to-earth, innovative company culture.
• Stebby wellness benefit, additional vacation and health days.