# Katana Healthcare — AI/ML Lead Engineer

| Field | Value |
|---|---|
| **Date found** | 2026-05-25 |
| **Company** | Katana Healthcare |
| **Role** | AI/ML Lead Engineer |
| **Location** | Ireland — open to fully remote (prefer Cork hybrid) |
| **Salary** | Undisclosed |
| **Job URL** | https://www.linkedin.com/jobs/view/4413694519/ |
| **Status** | New |

---

## Company Research

| Field | Value |
|---|---|
| **Headquarters** | Cork, Ireland |
| **Founded** | ~2022 (exact year not disclosed) |
| **Employees** | 11–50 (4 on LinkedIn) |
| **LinkedIn** | https://www.linkedin.com/company/katanahealthcare/ — 257 followers |
| **Website** | Not publicly listed |
| **Blog** | — |

- **Product:** AI-powered clinical decision support tools for frontline healthcare staff — fast, accurate, AI-driven information delivery to clinicians.
- **Customers:** Healthcare providers; tools validated through real-world clinical pilots.
- **Notable:** Very early-stage; small founding team; received external pilots. Focus on healthcare AI safety and accuracy.

---

## Job Summary

**What they do:** Katana Healthcare builds clinical decision support tools using AI to bring accurate, relevant information to frontline clinicians quickly.

**The role:** Lead AI/ML Engineer taking full ownership of the AI backend architecture, scaling RAG pipelines and LLM workflows from validated pilot stage to production.

**Core work:**
- Architect and deploy scalable RAG pipelines and LLM workflows for clinical decision support
- Integrate and fine-tune healthcare foundation models (MedGemma, Claude Healthcare, Gemini, Qwen, OpenAI)
- Build multi-modal data processing systems and risk probability algorithms over patient data
- Lead end-to-end MLOps lifecycle with clinical safety and regulatory compliance guardrails

**Stack:** Python · RAG · LLMs · MedGemma · Claude Healthcare · Hugging Face · MLOps · multi-modal AI

**Work style:** Ideally Cork hybrid but open to fully remote for the right candidate. Small high-trust team, startup pace.

---

## Score: 63%

| Dimension | Score | Justification |
|---|---|---|
| Agentic AI depth (25%) | 55% | RAG + LLM workflows prominent; agentic frameworks listed as "preferred" experience but not core. Primary focus is clinical ML and retrieval, not orchestration. |
| Tech fit (25%) | 65% | RAG, LLM fine-tuning, MLOps, Python — strong overlap. But stack centres on healthcare-specific models (not Anthropic/LangGraph primary). |
| Remote fit (25%) | 65% | "Open to fully remote for the right candidate" — positive signal, but preference for Cork hybrid creates uncertainty. |
| Company culture fit (15%) | 68% | AI-native healthcare startup, tiny team, genuine clinical impact, autonomous IC role. Domain is niche but mission-driven. |
| IC/leadership balance (10%) | 72% | "Lead the next phase of our platform" — predominantly hands-on technical ownership, not people management. |
| **Final (weighted)** | **63%** | Good fit on culture and IC ownership; moderate on agentic depth and remote certainty. |

---

## Strengths

- AI-native startup in healthcare: clinical domain overlaps with Luca's compliance/document AI background (Fenergo KYC work)
- RAG pipeline and LLM experience is a direct fit
- Founding/lead AI engineer role with full ownership over architecture
- Small team, high autonomy — matches preferred culture

---

## Weaknesses & Risks

- ⚠️ Lead role — flagged per scoring rules
- Salary undisclosed at very early-stage startup — could be below €110k target
- Remote policy ambiguous — "open to remote" but prefers Cork hybrid
- Healthcare-specific domain may limit career mobility vs pure AI-native stack
- Very small company (4 on LinkedIn) — execution risk

---

## Suggestions

- Ask early about remote policy and salary to avoid wasted time
- Emphasise Fenergo KYC document extraction work (clinical/compliance overlap) and MLOps track record
- Highlight RAG pipeline experience and production deployment work

---

## Interview Tracker

| Stage | Date | Notes |
|---|---|---|
| Expired | 2026-06-14 | Posting expired — no longer accepting applications |
| Applied | | |
| Recruiter screen | | |
| Technical interview | | |
| Final round | | |
| Offer / Outcome | | |