Agentic Workflows
Summary
Developed multi-step reasoning systems using CrewAI & AutoGen.
Tahir Askarli
AI Software Engineer
Baku, AZ.About
AI Engineer with 4+ years of experience building and operating production-grade AI/ML systems, specializing in RL environments, LLM systems, and autonomous decision pipelines across aerospace, energy, and retail. Proven ability to design and deliver high-reliability MLE/SWE environments and diverse tasks in mission-critical, regulated domains, leveraging strong Python, Docker, and end-to-end engineering ownership to meet demanding throughput expectations.
Work
Baykar Technologies
|AI Software Engineer
→
Summary
Leads the development of production-grade RL environments and advanced perception systems for mission-critical UAV platforms, ensuring high reliability and compliance.
Highlights
Lead the development of sophisticated RL environments for UAV swarm coordination and autonomous decision-making, significantly enhancing operational capabilities for mission-critical systems.
Design and implement deep learning perception systems for advanced UAV platforms (TB2, TB3), optimizing real-time performance and reliability.
Optimize edge inference using TensorRT and ONNX for real-time deployment, boosting system efficiency and responsiveness.
Build mission-critical Computer Vision systems, including autonomous landing on naval platforms, achieving 99%+ reliability in high-stakes operational environments.
Ensure stringent compliance with production-grade system architecture and comprehensive documentation, maintaining high standards for reliability, scalability, and maintainability.
eiLink R&D; Center
|Data Scientist - AI/ML Research
→
Summary
Developed specialized AI systems for oil & gas signal processing and collaborated with BP to deliver critical engineering tools.
Highlights
Developed advanced AI systems for oil & gas signal processing in the ACG field, optimizing data analysis and contributing to operational efficiency.
Built and deployed anomaly detection models for real-time sand event detection, significantly reducing false positives and preventing critical system failures.
Delivered a Python-based visualization tool in collaboration with BP engineers for pressure transient analysis, providing critical insights and supporting decision-making.
Aimerch
|AI Engineer - Multimodal & LLM Systems
Remote, US
→
Summary
Developed and deployed advanced multimodal and LLM systems for retail AI, establishing robust evaluation and risk assessment frameworks.
Highlights
Built and deployed production-grade LLM, RAG, and Vision-Language systems, enhancing retail AI capabilities and driving improved data processing efficiency.
Designed and implemented complex multimodal pipelines, effectively combining vision, language, and structured data to deliver comprehensive AI solutions.
Developed and operationalized robust evaluation and risk assessment frameworks for production AI, ensuring system reliability and ethical deployment in high-stakes contexts.
Education
Educational Inof Azerbaijan
Data Science
Baku Higher Oil School
BSc
Engineering
Grade: 90/100
Istanbul Technical University
Exchange Program
Erasmus Exchange
DeepLearning.AI
Coursera
RL , Vision, LLM, NLP, ML, DL Specializations
Languages
English
Azerbaijani (Native)
Skills
RL & AI
PPO, Q-Learning, LLMs, RAG, Agent Systems, Computer Vision, Reinforcement Learning, Generative AI, Multimodal AI, Autonomous Decision Pipelines, Machine Learning Engineering.
Programming
Python (advanced), C++, SQL, MATLAB.
DevOps & Systems
Docker, CI/CD, FastAPI, TensorRT, ONNX, AWS, Azure, Production Mindset, Debugging, Reliability, Iteration Speed, MLOps, Scalable Pipelines, Performance Optimization, Distributed Systems.
Frameworks
PyTorch, TensorFlow, Keras, Scikit-learn, HuggingFace, LangChain, CrewAI, AutoGen.
Other Skills
Product Ownership, Engineering Ownership, System Architecture, Documentation, Risk Assessment, Evaluation Frameworks, Data Processing, Anomaly Detection.
Projects
Cognitive-Emotional Modeling
Summary
Research on EEG-based multimodal emotional representations.
AI HR Avatar System
Summary
Real-time AI interviewer with speech, vision tracking, and behavioral analytics.
RL Environment Design
Summary
Built Simulation Environments for training Autonomous UAVS for dog fighting scenarios.