AI & Data Engineer (all genders)

About the Role

You'll be our second AI & Data Engineer, the Tbilisi counterpart to our Lisbon seat. The role is deliberately combined: agentic systems require modeled, governed data behind them, so you'll own both.

Day-to-day you'll ship production agent systems (tool-calling LLMs, RAG pipelines, evals, deployment, observability) and the data plumbing that feeds them (extraction, transformation, modeling, dbt, warehouse).

The bar at 12 months is 2+ production agentic AI systems used by enterprise clients, plus the internal HubSpot data warehouse layer that the broader practice and future engagements stand on.

In Phase 1, this role is internal-facing (no direct client contact). Client-facing engagement opens in Phase 2 once a 12-month delivery track record is documented.

What You'll Do

• You'll ship 2+ production agentic AI systems used by enterprise clients (deployed, signed-off via the Lisbon or DACH frontstage, in active use)

• You'll build and maintain our internal HubSpot data warehouse layer: dbt models, source freshness checks, tested transformations, usable by AI engagements and internal analytics

• You'll contribute to our reusable agent framework (tool-calling layer, prompt management, evals harness, observability) used by future AI Engineers across all sites

• You'll establish a documented evals and data quality bar: every shipped agent gets documented evals, drift monitoring, and a rollback plan; every shipped pipeline gets tests and freshness SLAs

• You'll maintain ≥ 65% billable utilisation on internal AI projects or client engagements via the Lisbon/DACH frontstage from month 6 onward

• You'll co-author 1+ public-facing technical asset (blog post or webinar) on our AI engineering or data engineering approach

• You'll contribute to AI proposal and scoping conversations as the technical authority alongside Sales/Advisory via the Lisbon or DACH frontstage

What You Bring

Must-haves

• 3 to 6 years shipping production AI/data systems beyond notebooks. You can walk us through what you shipped, what broke, and how you fixed it

• Direct production experience building agent systems with tool-calling LLMs (Claude, GPT, Gemini): tool schemas, multi-step reasoning, error handling, retries, observability

• Production data engineering with dbt (or comparable): modeled CRM or B2B data into a clean dimensional layer, operated a warehouse (BigQuery, Snowflake, Postgres, DuckDB) under change control

• Production RAG with vector stores (Pinecone, Weaviate, pgvector), hybrid search, re-ranking, and chunking strategies that work for enterprise documents

• Evals discipline: you've built automated eval suites (golden datasets, LLM-as-judge, regression tests) and dbt-style data tests, and treat both as production code

• Strong Python or TypeScript with tests, types, CI, deployment in cloud (AWS, GCP, or Azure), Docker, basic IaC

• Business-fluent English (daily working language across all sites)

• Tbilisi-based or willing to relocate, with sustainable 5-hour CET overlap commitment

Nice-to-haves

• Direct HubSpot APIs experience (REST, GraphQL, webhooks)

• Public open-source contributions to AI or data engineering tools

• Experience integrating AI systems with CRMs, ticketing tools, or warehouses at enterprise scale

• Experience as a hub or site foundation engineer

Your Team

You'll report directly to the Director Development & Technology, with day-to-day pairing with the Lisbon AI & Data Engineer and the Tbilisi Senior Backend Anchor. You'll join the AI Team as the Tbilisi seat, working closely with the Lisbon hub for AI delivery and across Tbilisi for data plumbing.

Tech Stack

• Languages: Python and TypeScript

• Agent frameworks: LangGraph, LangChain, CrewAI, Pydantic AI, Mastra (we pick per engagement)

• LLMs: Anthropic Claude (we are an Anthropic Build Partner), OpenAI, Gemini, with self-hosted options where the engagement demands it

• Data engineering: dbt, BigQuery / Snowflake / Postgres / DuckDB, Airbyte / Fivetran (where it fits)

• Vector / RAG: Pinecone, Weaviate, pgvector

• Observability: LangSmith, Helicone, Langfuse, Arize

• Cloud and DevOps: AWS, GCP, Azure, Docker, GitHub Actions

• CRM and integration: HubSpot APIs (REST, GraphQL, webhooks)

• Collaboration: Jira, Confluence, Forecast, Claude AI, Claude Code, Cursor

What We Offer

• Employment & Statutory: Compliant Georgian employment via Remote.com EOR (paid time off, public holidays, parental leave, statutory health coverage), salary paid in EUR

• Health: Private health insurance top-up on top of statutory coverage

• Learning: Full Blinkist Business library (4,500+ books), 3 months of Babbel, dedicated AI and data engineering certification budget

• Flexibility: Remote-first setup with Tbilisi hub access, daily CET overlap (5 hours minimum)

• Culture & Tools: Flat hierarchies with direct line to Director D&T and Leadership, modern stack (HubSpot, Jira, Confluence, Claude AI), Anthropic Build Partner status with early access to Claude capabilities

About Thorit

Thorit is one of Europe's leading HubSpot partners and a top business and technology consultancy. We combine strategic advisory with hands-on IT implementation, from CRM architecture to fully automated go-to-market infrastructure powered by AI and agentic systems.

Ready to Apply?

Send your application via Ashby. A cover letter is optional, but two artefacts are required: (1) a concrete example of a production agent system you have personally shipped (architecture, tooling, evals, rollout, what broke and how you fixed it), and (2) a concrete example of a production data pipeline you have personally designed and shipped (source systems, modeling decisions, tests, freshness, who consumed it). Public links (GitHub, blog post, demo video) very welcome.