Бiба

Profile
Yevhen Biba
Senior Full Stack Engineer with 5+ years building enterprise back-ends on C# / .NET 8 and
ASP.NET Core and shipping production front-ends in Angular, React and Vue. Designs
S E N I O R F U L L S TA C K E N G I N E E R distributed, event-driven services running on Kubernetes across Azure and AWS, backed by
· .NET / AI
SQL Server, Postgres, MongoDB, Redis and Kafka. Over the last two years focused on LLM-
powered product features — RAG pipelines, dataset preparation, fine-tuning and evaluation
Details — and on making AI a day-to-day part of the engineering workflow. Track record of cutting API
latency ~25%, lowering infrastructure cost through cloud migrations, and mentoring engineers
Mykolaiv, Ukraine into faster, safer delivery.
Remote / EU time zones
[open contact info](look above in the "contact info" section)
[open contact info](look above in the "contact info" section) AI & LLM Engineering
Product AI. Built LLM-backed features end to end on .NET: retrieval-augmented (RAG)
Skills search and assistant flows via Semantic Kernel and the Azure OpenAI SDK over Azure AI
Search and Postgres/pgvector, streamed to the UI over SSE/SignalR, with function calling
L A N G UA G E S
for structured actions and guardrails for PII and prompt injection.
C#, TypeScript / JavaScript, SQL, Go,
Python Datasets. Owned the data side — collected and normalised domain corpora and support
B A C K- E N D transcripts, de-duplicated and chunked them, wrote labelling guidelines, ran annotation
.NET 8, ASP.NET Core, Entity rounds and built synthetic-data generation to cover thin classes; produced clean
Framework, Minimal APIs, gRPC, train/validation/eval splits.
REST, GraphQL, SignalR, Blazor
Fine-tuning & adaptation. Fine-tuned hosted models on Azure OpenAI and open-weight
F R O N T- E N D models (LoRA/QLoRA on Llama- and Mistral-family) for domain tone and structured
Angular, React, Vue, Next.js, RxJS, output; compared them against prompt-only and RAG baselines on cost, latency and
HTML/CSS, Tailwind
accuracy before rollout.
AI / LLM
Evaluation & ops. Built regression eval suites (golden sets, LLM-as-judge, groundedness
Azure OpenAI, OpenAI, Anthropic
APIs, Semantic Kernel, RAG, Azure AI
and hallucination checks) into Azure DevOps pipelines, plus token/cost dashboards,
Search, pgvector, embeddings, LoRA / semantic caching in Redis and model fallback routing that kept inference spend
QLoRA fine-tuning, dataset predictable.
preparation & labelling, LLM
AI in the workflow. Daily use of Claude Code, GitHub Copilot and Cursor for
evaluation, Ollama / local LLMs,
Claude Code, Copilot, Cursor implementation, test generation, refactors and code review; wrote the team's
prompt/agent playbooks and local-LLM (Ollama) setup for work that could not leave the
DATA
network.
MSSQL, PostgreSQL, MongoDB, Redis,
Elasticsearch
MESSAGING
Kafka, RabbitMQ, Azure Service Bus
Experience
C LO U D
Senior .NET Developer, Amconsoft
Azure (AKS, App Service, Functions,
AUG 2025 — PRESENT
Azure SQL, Blob, Service Bus), AWS
(EKS, ECS, Lambda, S3, RDS, SQS) Lead full-stack delivery of enterprise applications on C# / ASP.NET Core with an
DEVOPS Angular/React front-end, shipping high-quality features on aggressive timelines.
Kubernetes, Docker, Helm, Terraform, Designed and shipped the platform's AI assistant: RAG over Azure AI Search and pgvector,
Azure DevOps, GitHub Actions, GitLab
tool/function calling into internal .NET APIs, streaming UI, and an eval harness that gates
CI
every prompt or model change in the pipeline.
O B S E R VA B I L I T Y
Prepared and curated the training and evaluation datasets behind it — cleaning, de-
OpenTelemetry, Application Insights,
Prometheus, Grafana, structured duplication, labelling guidelines and synthetic augmentation — and fine-tuned models
logging (Azure OpenAI and LoRA on open weights) for domain-specific extraction.
PRACTICES Cut inference cost with semantic caching in Redis, prompt compression and routing cheap
Microservices, event-driven design, traffic to smaller models, while keeping answer quality flat on the golden set.
DDD, Clean Architecture, TDD, code
Ran services on Kubernetes (AKS, Helm, HPA, rolling deploys); optimised database access
review, Agile / Scrum
and caching to keep p95 latency stable as load grew.
Raised engineering standards — unit and integration testing, CI gates, design reviews —
Languages
and mentored engineers across teams.
English — Upper-Intermediate
Ukrainian — Native

.NET Developer, Luxoft
JAN 2024 — JUL 2025

Built microservices in C# and ASP.NET Core integrating SQL Server, MongoDB, Redis and
third-party APIs, communicating over gRPC and Kafka / RabbitMQ.
Led migration of key services to Azure (AKS, App Service, Azure SQL) and containerised the
estate with Docker and Helm, improving availability and reducing infrastructure cost.
Delivered an internal document-understanding service: OCR + LLM extraction into
structured JSON, with a human-in-the-loop review UI that fed corrections back into the
training set.
Automated CI/CD in Azure DevOps, cutting manual deployment time and increasing release
cadence.
Improved observability with OpenTelemetry, Application Insights and Grafana dashboards,
reducing mean time to recovery.
Worked with product owners and QA in Agile ceremonies, contributed to technical design
and mentored peers.

Middle .NET Developer, GlobalLogic
FEB 2021 — DEC 2023

Developed and maintained REST APIs and back-end services in C#, ASP.NET Core and Entity
Framework; optimised data access and endpoints to reduce average API response time by
~25%.
Implemented front-end features in Angular and Blazor, improving user workflows and
component reusability.
Designed and tuned MSSQL schemas, queries and migrations to support growing data
volumes; introduced Redis caching for hot read paths.
Ran CI/CD pipelines and deployments in Azure DevOps and Azure App Service / AKS,
increasing release frequency and reliability.
Collaborated with product and QA in Agile sprints, performed code reviews and mentored
junior developers.

Education
Bachelor's Degree, Computer Science
B L A C K S E A N AT I O N A L U N I V E R S I T Y, M Y KO L A I V · S E P T 2 0 1 9 — J U L 2 0 2 3

Winner, BSNU startup contest 2020
Winner, BSNU hackathon 2021