Feliks Bańka

Feliks Bańka

AI Engineer & Researcher

About Me

I am a master’s student in Artificial Intelligence at Warsaw University of Technology with a strong foundation in machine learning, multi-agent systems, and game theory applied to AI. I currently work as an AI Engineer at AI Clearing, building production RAG systems and multi-agent architectures for complex analytical workflows. I have a growing track record of academic publications, including recent submissions to top-tier venues like ICML and DSS, as well as accepted papers at AAMAS (EXTRAAMAS), ICCS, PACIS (A-tier), and ACIIDS (Best Student Paper Award). I am passionate about creating robust, verifiable, and governance-aware AI systems.

Education

Master's degree • Warsaw University of Technology

Sep 2025 – Feb 2027

Major: Artificial Intelligence

Thesis: Investigating the Application of Game Theory in LLM-based Multi-Agent Systems

B.Eng. • Warsaw University of Technology

Sep 2021 – Jun 2025

Major: Automatic Control & Robotics

Thesis: AI Architectures for Option Pricing and Portfolio Management: Transformers to MAS

Publications

On Credit Assignment in Multi-Agent Reinforcement Learning using Hierarchical Credit-Conscious Actor-Critic

ICML 2026 (Under Review)

Proposes a hierarchical critic architecture for centralized training with decentralized execution that explicitly separates team-wide rewards from individual agent contributions. Validated on SMACv2 environments.

Beyond Accuracy: Structured Deliberation for Intelligent Decision Systems

Decision Support Systems (Under Review)

Investigates how multi-agent LLM deliberation can be organized as a governance-aware decision-support process. Evaluated using hedge-fund portfolio allocation, demonstrating stronger downside protection during extreme regime shifts.

Toward Explanatory Equilibrium: Verifiable Reasoning as a Coordination Mechanism under Asymmetric Information

AAMAS 2026 (EXTRAAMAS workshops)

Introduces Explanatory Equilibrium as a design principle for explanation-aware MAS. Models LLM communication as a signaling game to derive validation bounds preventing cheap-talk degeneration.

MultiHedge: Adaptive Coordination via Retrieval-Augmented Control

ICCS 2026

Proposes a hybrid architecture where an LLM produces structured allocation decisions conditioned on retrieved historical precedents, and execution is grounded in canonical option strategies.

DeltaHedge: A Multi-Agent Framework for Portfolio Optimization

PACIS 2025

Accepted to PACIS (A-tier per CORE). Presents a multi-agent RL framework that ensembles options-based hedges to improve risk-adjusted returns and stabilize portfolios in volatile markets.

Options Pricing Platform with NNs, LLMs & RL

ACIIDS 2025 • Best Student Paper Award

Integrated neural networks, volatility modeling, and sentiment analysis via LLMs into a hybrid option pricing model. Designed an RL trading strategy.

Transformer Architectures for Option Pricing and Forecasting

LNAI (extended ICAART)

Compares Informer, Autoformer, FEDformer, and Pyraformer against classical models and deep sequence baselines across equities, indices, and crypto options.

Applying Informer for Option Pricing

ICAART 2025, Porto

Explored the Informer—a Transformer-based model—for option pricing, adapting it to handle long-sequence modeling capabilities to enhance prediction accuracy.

Experience

AI Engineer • AI Clearing

Jul 2025 – Present

Architected and deployed a production multi-agent LLM system (LangGraph, LangChain). Designed a domain-specific RAG pipeline handling complex construction data (Excel, nested tables, marked-up PDFs) via OCR, vision LLMs, BM25, and reranking.

Impact: Increased system accuracy from 78% to 98%. Reduced latency (80s → 20s) via caching and token usage (1.1M → 200k) via selective tool-based routing.

Developer • Regionalne Przysmaki

Oct 2023 – Jun 2025

Developed Python-based automation tools and data pipelines. Analyzed market trends and optimized strategies to support business operations.

Full Stack Developer • Freeport Metrics

Jul 2023 – Oct 2023

Built React (TypeScript) and .NET micro-frontends, implemented REST APIs, and integrated backend logic in cross-functional Agile teams.

Extracurricular Activities

Co-founder & President • KN Intelligent Information Systems Nexus

Jun 2025 – Present

Leading a student research group focused on generative AI, multi-agent systems, LLM applications, and argument modeling. Promoting practical AI applications in finance and decision systems.

Vice President • KN Business in Technology

Feb 2023 – Jun 2024

Led AI/LLM-oriented student projects, organized tech-business meetups, and managed a 10-member team delivering workshops and hackathons in Warsaw.

Member • Enactus WUT

Nov 2021 – Sep 2024

Collaborated on entrepreneurship projects addressing social challenges through innovation, with emphasis on leadership and cross-functional teamwork.

Honors & Awards

EU / Ministry of Science Research Grant

FERS 2021-2027

Received research funding to support the development of multi-agent architectures and cognitive systems applied to intelligent information networks.

Minister of Science Scholarship

Minister Nauki i Szkolnictwa Wyższego (Poland)

Awarded for outstanding academic achievements and continued scientific development in the area of cognitive architectures and MAS in intelligent systems.

Best Student Paper Award

ACIIDS 2025

Honored for the paper "Options Pricing Platform with Neural Networks, LLMs and Reinforcement Learning" at the Asian Conference on Intelligent Information and Database Systems.

Skills & Tools

Machine Learning & AI

PyTorch Transformers Multi-Agent Systems Reinforcement Learning RAG Statistical Learning Theory Game Theory Probabilistic Modeling LangChain LangGraph OpenAI API Hugging Face

LLM Optimization & Data

Speculative Decoding Knowledge Distillation Quantization Logit Processing Vector DBs (Qdrant) Knowledge Graphs (Neo4j) pandas NumPy scikit-learn

Engineering & Infrastructure

Python TypeScript FastAPI Docker Kubernetes (K8s) Distributed Training Caching & Resource Opt. React.js AWS Azure DevOps MLflow

Languages

Polish • Native English • Advanced German • Intermediate

Hobbies & Interests

⚽️
Football Enthusiast
🎵
Songwriting & Music
📈
Stock Market Analysis
🤖
DIY Robotics Projects