We are seeking a hands-on AI architect and developer who designs, builds, tunes, and deploys production-grade machine learning and generative AI systems. Working as a forward-deployed engineer, this person sits close to the business: partnering directly with sales, partner, and customer success stakeholders to turn high-value go-to-market problems into reliable AI solutions that measurably drive consumed ACV (cACV) growth by accelerating customer adoption and consumption of cloud solutions.
The role combines three core disciplines: classical and deep machine learning, AI model tuning and optimization, and the architecture of AI harnesses: the orchestration, evaluation, tooling, and guardrail layers that make models dependable in real enterprise workflows. The role has a primary focus on data modeling and architecture: designing and maintaining conceptual, logical, and physical data models using dimensional, star/snowflake, Data Vault, and lakehouse medallion approaches; integrating SAP and non-SAP GTM data; and building a trusted semantic layer that supports analytics, ML features, and AI agents.
Responsibilities Design, train, and deploy ML models for prediction, classification, ranking, and recommendation use cases across the go-to-market lifecycle, including pipeline scoring, churn and expansion propensity, and opportunity prioritization Partner with data engineering to source, clean, and structure enterprise data, and build robust feature pipelines on Databricks and SAP Datasphere while ensuring data quality, lineage, and reproducibility Maintain conceptual, logical, and physical data models using dimensional, star/snowflake, Data Vault, and lakehouse medallion approaches Define hypotheses, success metrics, and offline/online evaluation methods, run A/B tests, and translate results into clear business recommendations Adapt foundation models through supervised fine-tuning, parameter-efficient methods, and preference-based techniques where appropriate Architect retrieval-augmented generation pipelines, embedding strategies, chunking, and re-ranking, and engineer and version prompts and system instructions as managed assets Optimize latency, throughput, and inference cost through model selection, distillation, quantization, caching, and routing between models of different sizes Benchmark commercial and open-weight models against internal tasks and recommend the right model for each use case based on quality, cost, risk, and compliance Architect the end-to-end AI harness that wraps models in production, including orchestration, tool and function calling, memory and context management, multi-step and multi-agent workflows, and fallback logic Build automated evaluation suites so every model or prompt change is measured before release Implement input/output validation, grounding checks, hallucination detection, PII handling, and policy enforcement aligned with enterprise security and responsible-AI standards Integrate AI solutions with enterprise systems via APIs, connectors, and secure data access patterns, and tie each solution to measurable business outcomes such as seller productivity, cycle-time reduction, win-rate improvement, and cACV contribution Requirements 5+ years of experience in machine learning engineering, with a track record of shipping models into production enterprise environments Qualifications in enterprise data platforms and lineage/reproducibility practices within large-scale data engineering environments Expertise in statistical experimentation, including hypothesis design and A/B testing methodologies for business-critical decisions Proficiency in working with foundation models and parameter-efficient adaptation techniques at scale Showcase of prompt and asset versioning practices within managed, auditable systems Knowledge of cost-performance trade-offs across inference infrastructure, including latency and throughput engineering Background in benchmarking methodologies for comparing model quality, risk, and compliance across vendors Familiarity with observability practices for AI systems, including tracing, logging, and drift monitoring in production Competency in enterprise security and responsible-AI compliance frameworks Understanding of how to translate technical AI outcomes into business metrics such as productivity, cycle time, and revenue impact Hungary Dynamic, entrepreneurial corporate environment Diverse multicultural, multi-functional, and multilingual work environment Opportunities for personal and career growth in a progressive industry Global scope, international projects Widespread training and development opportunities Access to LinkedIn Learning solutions Competitive salary and various benefits Advanced wellbeing and CSR programs, recreation area
Job description provided by EPAM Systems.
View the original posting