AI Lead Developer

Requirements:
Languages: Python (primary), production-grade, tested, reviewable.
Good to have: Bash/shell scripting, SQL.
C++ or Rust for inference optimisation good to have.
Must have:
Core ML frameworks: PyTorch (deep, hands-on).
Hugging Face Transformers, PEFT, TRL, Datasets, Accelerate, Tokenisers.
NumPy, SciPy, pandas, scikit-learn.
JAX / TensorFlow: good to have.
Must have:
Fine-tuning techniques: full fine-tuning and parameter-efficient fine-tuning LoRA, QLoRA, adapters, prefix tuning.
SFT (supervised fine-tuning) and instruction tuning.
Preference optimisation: DPO, ORPO, KTO, RLHF / PPO.
Continued pre-training / domain-adaptive pre-training.
Learning-rate scheduling, warm-up, gradient accumulation, gradient clipping.
Mixed precision (bf16 / fp16), gradient checkpointing.
Mitigating catastrophic forgetting, overfitting, and distribution shift.
Curriculum design and data mixture/sampling ratio tuning.
Checkpoint selection, early stopping, and model merging (SLERP, TIES) are good to have.
Must have:
Distributed training: DeepSpeed (ZeRO stages) and/or PyTorch FSDP.
Multi-GPU and multi-node training.
NCCL basics, sharding strategies, OOM debugging.
Flash Attention, xFormers, Liger Kernel good to have.
Must have:
Speech / ASR: Whisper, wav2vec2 HuBERT, Conformer, RNN-T / transducer models.
CTC vs. attention-encoder-decoder vs. transducer architecture trade-offs.
Domain adaptation: vocabulary biasing, contextual biasing, shallow/external LM fusion.
Forced alignment; pronunciation and lexicon handling for clinical and drug terms.
Audio pre-processing: feature extraction (log-mel, MFCC), VAD, resampling, normalisation.
Augmentation: SpecAugment, noise, reverberation, speed/pitch perturbation.
Speaker diarisation and speaker separation (pyannote or equivalent).
WER / CER evaluation sliced by accent, speaker, cohort and clinical term.
Streaming / low-latency ASR inference.
Toolkits: NeMo, ESPnet, SpeechBrain, Kaldi, torchaudio, librosa.
Must have:
LLM and Clinical NLP fine-tuning: open-weight LLMs (Llama, Qwen, Mistral, Gemma or equivalent) for a specialised domain.
Clinical NER and structured extraction from unstructured notes.
Terminology normalisation: SNOMED CT, ICD-10 LOINC, RxNorm, UMLS.
Handling failure modes: hallucination, omission, sycophancy, long-context degradation.
RAG and agentic pipelines; embedding models and vector stores.
Prompt engineering and structured/constrained output generation.
Clinical document standards: HL7 FHIR, SOAP notes good to have.
Must have:
Data Engineering: large-scale dataset curation, cleaning and de-duplication.
Labelling strategy, annotation guidelines and inter-annotator agreement.
Synthetic data generation and quality filtering.
Train/validation/test splitting with leakage detection.
Audio and text data pipelines at scale (Spark, Ray, Dask or equivalent).
Evaluation: Designing task-specific eval suites and regression harnesses.
Speech metrics: WER, CER, entity-level and keyword recall.
Generation metrics: rubric-based grading, LLM-as-judge (with its caveats), factuality and omission scoring.
Human/clinician review protocols and annotation tooling.
Statistical significance, confidence intervals, and per-cohort error analysis.
Bias and fairness testing across accent, gender, age and language.
Must have: Inference and Optimisation: Quantisation: INT8 / INT4 GPTQ, AWQ, bitsandbytes.
Knowledge distillation:
Export and runtimes: ONNX, TensorRT, CTranslate2
Serving stacks: vLLM, Triton Inference Server, TorchServe, TGI.
KV caching, continuous batching, latency and throughput profiling.
Infrastructure and MLOps:
Cloud GPU compute: AWS / GCP / Azure; SLURM or Kubernetes-based scheduling.
Docker and containerised training/serving.
Experiment tracking: Weights and Biases or MLflow.
Model and dataset versioning, artefact registries.
CI/CD for ML, automated retraining, monitoring and drift detection.
Git and code review discipline.
Cost awareness and GPU budget management.
Must have:
Domain and Compliance: Healthcare / medtech or another regulated domain.
PHI / PII handling, de-identification, HIPAA / GDPR / DPDP.
Clinical validation and SaMD / CE marking / FDA pathways.
Model documentation, audit trails and reproducibility for regulated release.
Leadership and Soft Skills:
Mentoring 3-5 engineers on applied AI practice.
Running design and code reviews; setting technical standards.
Translating clinical and product requirements into model objectives.
Explaining model behaviour, uncertainty and limitations to non-technical stakeholders.
Written communication: design docs, eval reports, post-mortems.
Pragmatic prioritisation and comfort with ambiguity.
Must have:
Credentials: 8-10 years in software / ML engineering, with 3-4+ years training or adapting deep learning models in production.
Publications, open-source contributions or benchmark results in speech or clinical NLP good to have.
Multilingual or code-switched speech modelling good to have.

Experience
8-10 yrs