Machine Learning Engineer, Researcher

Intellectsoft

. Drive end-to-end adaptation of small language models to specialized domain tasks, selecting base models, training strategies, and architectures.

Posted 10/28/2026full-timeBengaluru • IndiaMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

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

Demonstrates expertise in fine-tuning transformer-based language models and implementing parameter-efficient methods, with a strong command of the Hugging Face ecosystem and modern inference frameworks. Capable of building robust training pipelines and translating cutting-edge ML research into practical applications.

Highest-signal resume keywords

Machine Learning EngineeringFine-Tuning Transformer-Based ModelsAdvanced Python ProficiencyDeep Fluency in Hugging Face EcosystemExperience with Inference Frameworks

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-TuningParameter-Efficient MethodsSynthetic Data GenerationModel QuantizationTraining Pipeline DevelopmentExecutable Correctness EvaluationsData CurationLoRAQLoRAReinforcement Learning

Tools & Technologies

PyTorchHugging Face TransformersPEFTDatasetsAccelerateTRLVLLMTGIOllamaTensorRT-LLM

Industry Keywords

Machine LearningNatural Language ProcessingModel DistillationContinued Pre-TrainingReinforcement LearningBenchmarkingData FormattingTokenizationDistributed TrainingEdge Deployment

Tech Stack

Tools & technologies

PythonPyTorchSQL

About the role

Key responsibilities & impact

Drive end-to-end adaptation of small language models to specialized domain tasks, selecting base models, training strategies, and architectures.
Implement parameter-efficient fine-tuning methods with a mathematically grounded understanding of LoRA, QLoRA, rank selection, parameterization, scaling, and accuracy–efficiency trade-offs.
Determine and sequence adaptation strategies including Continued Pre-Training, Supervised Fine-Tuning, and RL/preference tuning such as DPO, GRPO, and RLHF.
Clean, curate, format, and generate high-quality datasets, including synthetic data generation when domain data is scarce.
Build task-specific benchmarks grounded in functional, executable correctness and maintain regression tracking across versions.
Use larger models as teachers to transfer capabilities to small language models via distillation, structured pruning, and quantization.
Stand up and optimize inference infrastructure across GPU, CPU, and edge targets using frameworks such as vLLM, TGI, Ollama, llama.cpp, or TensorRT-LLM.
Maintain reproducible pipelines with data/config versioning, checkpointing, and experiment tracking.
Break target tasks into milestones with clear criteria.
Translate recent literature into reproducible experiments.

Requirements

What you’ll need

3+ years in Machine Learning Engineering or Applied ML, with hands-on experience fine-tuning transformer-based language models.
Advanced Python and PyTorch proficiency.
Deep fluency in the Hugging Face ecosystem, including Transformers, PEFT, Datasets, Accelerate, and TRL.
Practical command of SFT with LoRA/QLoRA, alongside the judgment to reason about CPT and RL/preference methods.
Strong understanding of base model architectures, tokenizer design, pre-training data distributions, licensing, and context window requirements.
Proven experience building training pipelines, including curation, cleaning, synthetic data generation, and formatting.
Ability to design executable correctness evaluations, with hands-on experience serving and quantizing models using at least one modern inference framework.
Comfort reading recent ML research papers and translating state-of-the-art methods into working code.
Nice to have: Hands-on experience with Continued Pre-Training (CPT) on large domain corpora.
Nice to have: Experience with Reinforcement Learning / preference optimization for LLMs, including reward modeling, DPO, GRPO, or verifiable-reward setups.
Nice to have: Experience with structured-output generation, code generation such as SQL/DDL, or domain classification fine-tuning.
Nice to have: Familiarity with distributed training frameworks such as DeepSpeed and FSDP.
Nice to have: Knowledge distillation experience transferring capabilities from large teacher models to small student models.
Nice to have: Experience with on-device or edge deployment.
Nice to have: Active open-source contributions to machine learning tooling, libraries, or published models.

Benefits

Comp & perks

Employment-based cooperation
Comprehensive insurance for you and your family (health, life, and accident)
Paid PTO policy (vacation, sick leaves, and public holidays)
Tech equipment provided
Udemy courses, workshops, trainings & expert knowledge-sharing
Flexible hours & work setup