# AI

> AI-native development, agentic systems for ops & engineering

*1 year focused experience*

AI-native development approach grounded in a Master's in Data Science and Financial Technologies. Designed and deployed agentic systems to streamline operations and accelerate software development workflows.

## Highlights

- Apply AI-native practices across the full engineering lifecycle
- Design and deploy agentic systems that automate complex operational workflows
- Approach AI with a scientific rigour — Master's in Data Science underpins every model decision
- Bring a systems engineering lens to AI: reliability, observability, failure modes

## Stack

AI-native, Agentic Systems, LLM Integration, PyTorch, MLX, Fine-tuning, NLP, MLOps

## Scientific foundation

A Master's in Data Analysis and Financial Technologies from HSE means AI is not a buzzword — it's a discipline. Model selection, evaluation methodology, and statistical rigour are built into how I approach every AI problem.

## Agentic systems

Designed and deployed systems where AI agents take on real operational workloads — not demos, but production systems with real business impact.

- Autonomous agents for operations: alert triage, runbook execution, incident response
- Development workflow agents: PR review, test generation, architecture validation
- Multi-step reasoning pipelines with tool use, memory, and structured output
- Evaluation frameworks to measure and improve agent reliability over time

## The engineering lens

AI systems require the same engineering discipline as any other production system — reliability, observability, failure modes, and graceful degradation.

- Fine-tuned open-weights models matching commercial LLM quality at lower cost
- LLM integration patterns: streaming, tool use, structured output, caching
- Evaluation and feedback loops built into the deployment lifecycle
- Cost, latency, and reliability tradeoffs navigated with production constraints

## Relevant experience

### TradingView — Jan 2025 – Present

Principal Engineer · Team Lead

- Developing internal AI platform to accelerate product and software development lifecycle processes
- Designed and developed 2 new products end-to-end, expanding the company's portfolio
- Built an AI Skills Library embedding AI tooling into the SDLC; ran adoption workshops across engineering teams
- Prototyped AI-native products to streamline roadmap prioritisation
- Fine-tuned open-weights models for data enrichment pipelines, matching commercial LLM quality at lower cost

## Other areas

- [leadership](/leadership) ([markdown](/leadership.md))
- [backend](/backend) ([markdown](/backend.md))
- [devops](/devops) ([markdown](/devops.md))
- [frontend](/frontend) ([markdown](/frontend.md))

---

[Back to CV](/cv) ([markdown](/cv.md))