Index
TS · 2026
Location
Hamburg, DE
Affiliation
InnoGames · Leuphana Uni.
Available
April 2027
Last revision
30 August 2026
§ 00 · Abstract

LLM agent systems for analytics and forecasting.

AI Engineer and Data Scientist specialising in LLM agent systems for analytics workflows. At InnoGames, I develop an agentic time-series forecasting solution that combines foundation models such as TabPFN with an LLM agent to incorporate business planning information into forecasts. I also develop an agentic workflow that builds Power BI reports, implements report changes, and validates the results. I bring ten years of analytics experience across telecommunications, finance, insurance, and gaming.

10
years in analytics
1.9
M.Sc. GPA (Leuphana)
8.0
IELTS (English C1)
B2-C1
German
Fig. 1 · Author
Timur Salakhetdinov
name: timur salakhetdinov
profile: ai engineer & data scientist
edu:  leuphana m.sc. 2024–27
tz:   europe/berlin (cet)
§ 01 About
A brief CV

LLM agent systems for analytics, complemented by time-series forecasting with foundation models.

One of my current projects is an agentic time-series forecasting solution that combines foundation models such as TabPFN with an LLM agent, which turns versioned planning documents into forecasting covariates. My M.Sc. thesis at Leuphana University examines whether those text-derived covariates improve net revenue forecasts at InnoGames.

A second project is LLM-based tooling for analytics tasks such as building Power BI reports and preparing forecasting inputs. The workflows include explicit approval, validation, and escalation steps so that changes remain controlled and reviewable.

I am completing an M.Sc. in Management & Data Science at Leuphana University as a Deutschlandstipendium scholar, alongside my analytics and agent development work at InnoGames. From April 2027 I am available for full-time AI Engineer and Data Scientist roles in Germany.

PeriodRole / programmeType
Nov 2025 →
Working Student, Analytics · InnoGames
Hamburg, hybrid · Microsoft Fabric, Power BI, Hive SQL, StarRocks, Superset · TabPFN / PyMC / Vertex AI
Part-time
2024–27
M.Sc. Management & Data Science
Leuphana University Lüneburg · GPA 1.9 · expected Mar 2027
Study
2024–25
Data Analyst · Absolute Insurance
Moscow, remote · MS SQL · Power BI dashboards · clustering for underwriting
FT
2022–23
Data Science and Data Analysis Bootcamps
Yandex Practicum · applied ML, Python, SQL, statistics
Study
2019–23
Financial Analyst · Arkada Company
Moscow · telecom market research, valuation, competitor analysis
FT
2019–20
Professional retraining in Financial Analysis
HSE University, Moscow
Study
2011–19
Market Research Analyst · MTS PJSC
Moscow · international telecom markets (India, Serbia, Slovenia, Moldova)
FT
2006–11
Corporate Affairs Associate · Agency for Property Management
Moscow · analysis of company holdings, documentation for mergers and divestitures
FT
1999–05
Engineer-Economist, Economics & Management
RUDN University, Moscow · graduated with Distinction · KMK statement of comparability (2023): equivalent to German Master's level
Study
§ 02 Skills
self-assessed proficiency
Languages
Python (NumPy, Pandas, Scikit-Learn)
SQL (MS SQL Server, Hive, StarRocks)
LLM & Agents
LangGraph
Google ADK
Claude Code (agents, skills, plugins)
MCP
Tool calling
RAG
ML & Forecasting
TabPFN
Chronos
TimesFM
LightGBM
PyMC
PyTorch
Methods
Time-series forecasting
A/B testing
Cohort and churn analysis
Customer segmentation
Tools & Platforms
Microsoft Fabric
Power BI
Apache Superset
GCP (Vertex AI)
Docker
Git
Languages spoken /
EnglishC1 · IELTS 8.0
GermanB2-C1
RussianNative
§ 03 Projects

Selected projects: LLM agent systems, knowledge graphs, and Bayesian modelling

7 entries · work, open source & coursework

What I work on

My projects span LLM agent systems, time-series forecasting, Bayesian modelling, and knowledge-graph applications. They include a multi-agent toolkit used by an analytics team, a knowledge-graph question-answering system, a hierarchical Bayesian customer lifetime value model, and a multi-agent assistant for company founders.

IDPreviewTitle & abstractMethodsYear / Role
P-007
In daily use
PLAN APPROVE BUILD VERIFY CLEANUP

Agentic toolkit for Power BI report changes

Multi-agent Claude Code toolkit used across the analytics team, planning, implementing, and verifying Power BI dashboard changes in PBIP/TMDL without step-by-step human input.

Claude CodeMulti-AgentPBIP / TMDLPower BI
2026
Solo
P-006
ChefTreff AI

ChefTreff AI Hackathon: Founder Assistant

Multi-agent assistant that helps founders launch a company in Germany, covering business plan, marketing plan, and legal documentation.

Google ADKMulti-AgentLLM
2026
Hackathon · Hamburg
↗ ChefTreff hackathon
P-005
Coursework
Project thumbnail for a knowledge graph question answering system with SPARQL retrieval.

Knowledge Graph Question Answering

SPARQL-based question answering over structured knowledge graphs, enabling natural language queries over RDF knowledge bases.

SPARQLRDFNLP
2026
Team project
P-004
Industry partner
Project thumbnail for a machine learning route modelling project using KVG public transport data.

KVG ML Route Modelling

Applied ML to model and predict transportation routes for a regional transit partner, combining geospatial data with predictive modelling.

MLGeospatialForecasting
2026
Team project
P-003
Coursework
Project thumbnail for a Neo4j graph analysis of artist influence networks.

Neo4j Graph Analysis of Artist Influence Networks

Graph database solution using Neo4j and Cypher to analyse artists' influence and cluster musical lineages.

Neo4jCypherGraph
2025
Team project
P-002
Replication
Project thumbnail for a hierarchical Bayesian Pareto/NBD replication of Abe (2009).

Hierarchical Bayesian Pareto/NBD: Replication of Abe (2009)

Implemented and validated the hierarchical Bayesian Pareto/NBD model from Abe (2009) on the canonical CDNOW dataset.

PyMCMCMCBayesian
2025
Team project
P-001
ChefTreff AI

ChefTreff AI Hackathon: Product Detection

Built in 24 hours: a computer vision pipeline for identifying broken or damaged objects from product imagery.

PyTorchCNNCV
§ 04 Writing

Writing on LLM agent systems and applied machine learning

Latest from medium.com/@timursalakhetdinov
§ 05 Contact
open inbox · response within 24h

Available in Germany from April 2027, preferably Hamburg, Munich, or Düsseldorf.

I am interested in AI Engineer and Data Scientist positions involving LLM-based systems, forecasting automation, or analytics platforms. I combine hands-on AI development with ten years of experience translating business requirements into maintainable data and analytics solutions.

My work fits six kinds of teams well:

  • Insurance and reinsurance analytics. Underwriting segmentation at Absolute Insurance and current Bayesian forecasting research, applicable to Munich's insurance cluster and beyond.
  • Gaming and consumer products. Revenue forecasting prototypes and analytics infrastructure at InnoGames in Hamburg, including Microsoft Fabric migration and Power BI dashboards.
  • AI-native startups and applied AI labs working on LLM systems, agent architectures, and retrieval. I developed and rolled out a multi-agent toolkit now used daily by an analytics team, and my thesis develops an agent that converts planning text into inputs for revenue forecasting.
  • Telecommunications and network analytics. Eight years at MTS covering market analysis and M&A across the Indian, Serbian, Slovenian, and Moldovan markets, relevant to telecommunications operators and network businesses, including in Munich.
  • Global technology firms and platform teams in Hamburg, Munich, and Düsseldorf, where forecasting, applied ML, and production analytics scale across products.
  • Startups and scale-ups where production ML and agent systems form the core product rather than a pilot.

I work in English at C1 level (IELTS 8.0) and in German at B2-C1 level. My analytics experience spans international teams in finance, telecommunications, insurance, and gaming.