Piotr Bandurski

Software Engineer: AI agents and distributed systems

Summary

Software engineer with more than ten years of production experience, from Big Data platforms and event-driven Scala services to AI agents in Rust. I build backend systems end to end, from design and code to deployment and monitoring, and I enjoy leading the technical direction of a team.

Skills

Languages
Rust, Scala, Python, TypeScript, Go, C, R
AI and ML
LLM agents, OpenAI APIs, MLflow, Scikit-Learn, Keras, catboost, spaCy, gensim
Backend and data
Cats Effect, Akka, http4s, FS2, Kafka, Apache Spark, Apache Hive, Hadoop, PostgreSQL
Cloud and operations
AWS, GCP, Terraform, Kubernetes, Helm, ArgoCD, Docker, OpenTelemetry, DataDog, GitHub Actions

Experience

AI Engineer · Scalac

Jan 2026 – Present

Building an AI assistant that helps recruiters find and contact candidates.

  • Designed and implemented the core agent backend in Rust: agentic tool-use loop, multi-turn conversation state, PostgreSQL persistence
  • Made large result tables appear in milliseconds instead of seconds: the agent sends tables as structured data instead of having the LLM write them out token by token
  • Connected the assistant to a second, independent AI agent over the A2A (Agent-to-Agent) protocol, including a shared contract and live progress updates
  • Delivered agent features from design to production, such as sending outreach messages to candidates
  • Prototyped Agent-to-UI (A2UI) rendering, which lets the LLM build interactive views and custom components inside a chat interface
  • Built evaluation tooling for a non-deterministic agent: repeated A/B runs that measure the effect of a prompt or tool change, and a Rust proxy that runs the agent against a sandbox system
  • Implemented distributed tracing (OpenTelemetry) and LLM observability (MLflow GenAI traces) across the full agent stack
  • Used a spec-driven development workflow in which several AI coding agents implement and review each change

Rust · TypeScript · Python · Terraform · PostgreSQL · A2A · MCP · OpenTelemetry · MLflow · OpenAI APIs · Docker · Helm · ArgoCD · AWS · GitHub Actions

Senior Scala Engineer · Scalac

Wrocław · Apr 2023 – Dec 2025

Contributing to a modern, scalable recruiting and referral platform for a global HR technology company.

  • Acted as technical lead of a six-engineer team: set technical direction, reviewed most of the team's code and planned releases
  • Led a referral-link feature that spans three services, from design to a production launch without incidents
  • Set the Scala 3 stack and coding standards for the team's first two services (Cats Effect, Tapir, Doobie, hexagonal architecture); both are still in production
  • Built event-driven microservices with Kafka, including read models in PostgreSQL that are built from Kafka events
  • Traced a long-standing image-quality bug across four legacy services and shipped the fix with a safe data-migration and rollout plan
  • Mentored a junior engineer to independent work across services through gradual delegation and paired production deployments
  • Started the team's practice of architecture decision records (ADRs) and wrote shared Kafka guidance: an incident runbook and consumer benchmarks
  • Maintained CI/CD pipelines, cloud infrastructure (AWS, Terraform) and production monitoring (OpenTelemetry, DataDog)

Scala 3 · Cats Effect · Tapir · Doobie · Akka · Kafka · FS2 · Protobuf · OpenTelemetry · AWS (ECS, Lambda, S3) · Terraform · PostgreSQL · http4s · DataDog

Scala/Data Engineer · Scalac

Remote · May 2019 – Mar 2023

  • Healthcare backend (2021-2023): implemented a service for dynamic chat routing between client and healthcare insurance agents; refactored microservices to up-to-date Play/Play JSON libraries
  • Blockchain indexers for BTC, ETH and XTZ (2021): extended the ETH indexer to account for ERC-20 and ERC-721 token data
  • Automated data integration platform (2020-2021): extended the API and fixed production issues
  • Data Lake based on CDH 6.2.0 for a global company from the banking industry (2019-2020): Apache Hive and Hadoop, complex multi-stage Apache Spark jobs

Scala (Akka/Play, Typelevel) · Python · Apache Spark · Apache Hive · Hadoop · PostgreSQL · Kubernetes · GCP · Docker

Machine Learning Specialist · GetResponse

Gdańsk · Jan 2019 – Apr 2019

  • Designed and developed a Machine Learning development environment based on Jupyterhub and MLflow
  • Designed and developed a ML models productization pipeline based on MLflow, Docker, GitlabCI and Kubernetes
  • Designed and productionized a fraud detection algorithm based on catboost library
  • Developed a topic modeling solution for customer chat clustering based on LDA algorithm

Python · Scikit-Learn · Keras · catboost · spacy · gensim · Docker · Jupyterhub · GitlabCI · Kubernetes

Data Analytics Solutions Developer · Intel Corporation

Gdańsk · Apr 2017 – Dec 2018

Developing a Big Data System for products validation in Intel Data Center Group.

  • Took part in a startup initiative in DCG to collect and analyze Intel IPs telemetry and performance using Big Data & Machine Learning tools
  • Was responsible for Cloudera EDH Big Data Cluster and Apache Kafka cluster management & benchmarking
  • Developed a custom Kafka producer in Golang with support of Confluent Schema Registry and Apache Avro
  • Developed a number of Apache Spark applications using Spark SQL and Structured Streaming APIs

Python · Scala · Go · Cloudera EDH · Apache Hadoop · Apache Spark · Apache Kafka · Apache Oozie · Docker · Ansible

Firmware Engineer · Intel Corporation

Gdańsk · Feb 2015 – Apr 2017

  • Supporting development of Firmware for next-gen server platforms; took a role of build system product owner
  • Designed & implemented a data-driven solution that led to saving as much as 30 man-hours per week spent solely on FW compilation

Software Engineer Intern · Intel Corporation

Gdańsk · Mar 2014 – Jan 2015

Developing firmware for Intel server platforms.

Software Engineer Intern · Voicelab

Gdańsk · Oct 2012 – Feb 2014

Developing software for state-of-the-art automatic speech recognition system.

Education

Data Engineering - Data Science · Gdańsk University of Technology

2016 – 2017

M.Sc., Control Engineering and Robotics · Gdańsk University of Technology

2013 – 2015

B.Sc., Control Engineering and Robotics · Gdańsk University of Technology

2009 – 2013

Languages

  • Polish (native), English (full professional), Spanish (limited working), French (limited working), Japanese (elementary)

Certifications

  • Machine Learning; Deep Learning (5-course specialization by deeplearning.ai on Coursera); Introduction to Data Science; Text Mining: Bag of Words; The Data Scientist's Toolbox