Senior AI Researcher, LLM post-training – Hallucination Reduction

Aleph Alpha

. Design, execute, and scale SFT and RL post-training methodologies to reduce hallucinations and ground models in evidence .

Posted 10/14/2026full-timeHeidelberg • GermanySeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in fine-tuning and post-training methodologies for large language models, with a focus on hallucination mitigation and evidence grounding. Proficient in building and optimizing distributed training pipelines and evaluating model performance metrics.

Highest-signal resume keywords

Fine-Tuning LLMsPython ProgrammingPyTorch FrameworkDistributed Training ToolsHallucination Mitigation

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Fine-TuningPost-Training MethodologiesUncertainty EstimationCalibration MechanismsSelf-ConsistencySelf-CorrectionEvaluation BenchmarkingResearch ImplementationMulti-Node GPU ClustersRetrieval-Augmented Generation

Tools & Technologies

PyTorch FSDPDeepSpeedMegatron-LMTRLVerl/OpenRLHF

Industry Keywords

Large Language ModelsDistributed Training FrameworksTruthfulness AlignmentOut-of-Distribution DetectionClaim Verification

Tech Stack

Tools & technologies

Node.jsPythonPyTorch

About the role

Key responsibilities & impact

Design, execute, and scale SFT and RL post-training methodologies to reduce hallucinations and ground models in evidence
Research and implement uncertainty estimation, calibration, self-consistency, and self-correction mechanisms
Build and optimize high-throughput distributed fine-tuning and training pipelines across multi-node GPU clusters
Design and implement evaluations benchmarking hallucination rates, truthfulness, and reasoning accuracy
Independently drive research projects from hypothesis through scalable production implementation and technical reports
Collaborate with researchers and PhD interns across pre-training, mid-training, post-training, grounding, and specialized models

Requirements

What you’ll need

Hands-on experience fine-tuning or post-training LLMs using modern distributed training frameworks
Track record of research or implementation in hallucination mitigation, truthfulness alignment, grounding in RAG, uncertainty estimation, out-of-distribution detection, Retrieval-Augmented Generation evaluation, alignment, refusal behavior calibration, or claim verification
Strong programming skills in Python and PyTorch
Ability to travel to Heidelberg once per month
Experience with LLMs of 1B+ parameters
Hands-on experience with SFT, DPO or similar preference methods, PPO, or GRPO/other RL for LLMs
Experience configuring distributed training setups, including multi-node GPU clusters
Experience with distributed training tools such as PyTorch FSDP, DeepSpeed, Megatron-LM, TRL, or verl/OpenRLHF

Benefits

Comp & perks

30 days of paid vacation
Access to fitness and wellness offerings via Wellhub
Mental health support through nilo.health
Substantially subsidized company pension plan
Subsidized Germany-wide transportation ticket
Budget for additional technical equipment
Flexible working hours
Hybrid working model
JobRad® Bike Lease