← Усі вакансії

AI Engineer

Tricorder Systems
Формат:
повний remote
Зарплата:
$30 000 - $60 000 на місяць
Джерело:
jobs.dou.ua
Відгукнутись на вакансію →

We’re looking for a hands-on engineer to build the foundations of the Tricorder product: an offline AI scribe and decision support system for medics in austere environments.

Most healthcare technology assumes you’re in a hospital with reliable power and internet connectivity. However, care increasingly happens in the critical minutes before a patient reaches a ward — directly in the field, in transit or in remote and contested environments.

Tricorder builds the deployable clinical infrastructure for that world: a rugged, self-contained hardware and software platform that lets clinical teams capture, monitor and act on patient data anywhere, even when the network or the grid can’t be relied on. It is a dual-use venture serving defense medical and civilian pre-hospital care.

Our prototype product is already in user testing. Edge AI sits at the core: multimodal models that can turn video, sensor data and clinical notes into patient records, drive predictions and inform data-driven learning across deployments. To select, improve and evaluate those models in the field, we need rigorous data and evaluation — that’s where you come in.

What you’ll do

You’ll own two workflows, working day to day with the founder and the FF build team.

1. Edge AI Pipeline Development and Evaluation

Build the foundational architecture for the Tricorder product, focusing on offline data capture at the point of care.

Model selection: evaluate and select the best possible open source models for our use case. Explore architectures beyond LLMs that can run locally on the smallest possible RAM footprint.

Scoping custom fine-tuning: evaluate tradeoffs around fine-tuning off-the-shelf models, leaving them as is.

Dataset preparation: work with our CMO to assemble balanced, well-documented datasets, augment data where necessary or even generate synthetic examples for training and testing.

Eval harness: build an automated harness that benchmarks the system we’re developing against clinically meaningful metrics, tracks regressions and supports build-vs-buy model decisions.

2. Off-line decision support systems

Turn clinical guidelines into locally stored data structures that can be queried, digested and surfaced by our system.

Database design: design the schema and stand up the data store for clinical concepts and their relationships.

Database search methodologies: develop systems for dynamically surfacing clinical guidelines in response to real-time actions in the real world.

Document handling: parse and structure clinical documents (PDFs, notes, forms) into the database.

Clinical knowledge: work with clinicians to make sure the database reflects real-world clinical reasoning and pre-hospital workflows.

By the end of the first 3 months, we’d expect a working evaluation harness for guiding model selection and development, a labeled and versioned core dataset, and a first version of the decision support system linked to the product.

What we’re looking for

We don’t expect one person to be an expert in everything below. Strong candidates will be deep in one workstream and credible in the other.

Must-haves

Strong software engineering skills, with a track record of shipping production AI and ML pipelines, not just research notebooks.

Hands-on experience building ML evaluation frameworks, benchmarks or test harnesses, ideally for multimodal models.

Experience running data labeling and training-data preparation, including annotation tooling and quality control.

Working knowledge of databases and data modeling, embeddings, vector search.

Practical NLP experience: entity extraction, entity linking or document parsing.

Comfort handling sensitive data under data protection and clinical safety constraints.

Pragmatism and speed: you can work with ambiguity, make good build-vs-buy calls and ship in weeks, not quarters.

Nice to have

A passion for open source and edge AI.

A background in health-tech, clinical informatics or another regulated domain suc

Схожі вакансії

З блогу Trackr

Усі статті →

Знайдено через trackr.help/jobs · Канал: @trackrhelp · Бот для персональних сповіщень: @trackrhelpBot