← back to queue

ViyaMD | Staff Software Engineer, Staff ML Engineer | SF Bay Area | ONSITE ViyaMD is building a physician-grade clinical

ViyaMD · Remote · hn · fit score 25.0
Open / Apply ↗
⬇ Tailored résumé Queue →

Job description

ViyaMD | Staff Software Engineer, Staff ML Engineer | SF Bay Area | ONSITE ViyaMD is building a physician-grade clinical intelligence platform for legal, insurance, and healthcare teams. We analyze large, messy medical record sets and turn them into clinically accurate timelines, reasoning, and insights in minutes. We’re hiring for two senior roles: # Staff Software Engineer: Own core product and platform systems: medical record ingestion, document understanding pipelines, backend/data infrastructure, clinical workflow, and production reliability. We’re looking for: * 5–8+ years of software engineering experience * Strong Python/Typescript/backend engineering skills * Experience with distributed systems, data-intensive apps, or large-scale document processing * Ability to own ambiguous problems from architecture through production * Interest in healthcare, clinical data, legal-tech, or AI products # Staff ML Engineer: Own how ViyaMD evaluates and improves its clinical AI: define eval metrics, run model/agent evaluations, improve prompts and workflows, and close the loop through better retrieval, agentic systems, and fine-tuned models. We’re looking for: * 5–8+ years of professional experience * Strong hands-on AI/ML evaluation experience * Experience with LLMs, RAG, agents, fine-tuning, or applied ML systems * Fluency in Python * Healthcare or clinical experience is a strong plus, but not required We’re especially excited about people who want to set the technical bar, work c

✍️ Tailored application (cached — grounded in your resume)

58 / 100 honest fit

Good directional fit for the Staff Software Engineer posting: four years of Python backend services, AWS (Lambda/S3/DynamoDB) data pipelines, and Jenkins CI/CD at Medtronic on connected medical devices, plus real end-to-end ownership shipping and operating production Flask services on his own infrastructure using LLM/agentic tooling. The regulated-healthcare context and the correctness-first instinct are authentic, not aspirational. Honest gaps: ~4.5 years of professional experience against a 5–8+ year Staff bar, no TypeScript, no large-scale document-processing experience, and no formal ML evaluation / RAG / fine-tuning work — which makes the Staff ML Engineer posting a substantially weaker fit than the Staff SWE one; the SF Bay Area onsite requirement is also a move from Northridge.

Tailored resume highlights

"Why this company?" (draft — personalize before sending)

Two things pull me toward ViyaMD. I've spent four years at Medtronic writing software for connected medical devices, where correctness is a regulatory requirement and a wrong number can hurt someone — turning messy medical records into clinically accurate timelines carries that same weight, and I'd rather work where that's taken seriously. Second, I've been building with Claude Code daily for the past year, shipping production Flask services end-to-end on my own infrastructure. ViyaMD is applying agentic AI to a domain where being confidently wrong isn't acceptable, which is the harder and more interesting version of what I've been doing on my own.

Cover letter (review before sending)

Dear ViyaMD team, I'm applying for the Staff Software Engineer role. For the past four years at Medtronic I've built backend services and validation pipelines for connected medical devices — Python services, AWS (Lambda, S3, DynamoDB) data processing, and Jenkins CI/CD automation — in a regulated environment where a wrong output is a patient-safety problem, not a bug ticket. A physician-grade clinical platform has to hold that same bar. Much of my work has been root-cause analysis on messy, cross-team failures: reproducing rare field-found defects in-house and driving resolution across embedded and cloud components. Turning large, messy medical record sets into accurate timelines is a different domain, but a familiar shape — ambiguous inputs, a high correctness bar, and ownership from architecture through production. Separately, I've designed and operate a suite of production Flask applications on a self-hosted Linux server, built end-to-end with LLM/agentic tooling (Claude Code) — live services with hardened endpoints, rate limiting, cron data pipelines, and self-healing watchdogs. I'm also finishing an M.S. in Computer Science at Georgia Tech. I'd welcome the chance to talk about the ingestion and document-understanding side of the platform. Daniel Maynard

ATS keywords you have but should add

healthcare dataproduction reliabilityLLM / agentic systemsAI productsdata-intensive applicationsdata ingestion pipelinesdistributed systemsend-to-end ownership (architecture through production)FlaskLinux / self-hosted infrastructuredata infrastructure

Everything is grounded strictly in your resume — review before you submit. Nothing is auto-sent.