Where Agentic AI meets clinical practice — not to replace the clinician, but to restore the time, clarity, and joy that medicine was always meant to have. Built by a physician who got tired of fighting the technology.
These are the ideas that drive every line of code — written in clinic corridors, after long shifts, and during the quiet hours of building.
"We aren't just cutting down the clicks, we're elevating the decisions. The Quintuple Aim — in lines of code."
"Stop chasing the vibe. Start building the solace. Solace Coding is the intentional practice of getting under the hood — the quiet, focused hours where I find solace in the logic of Python and the architecture of LLMs."
"These agents don't just automate the discharge process — they master it. While the clinician focuses on the patient, the agents are running deep history synthesis, SDOH integration, guideline integrity, and predictive risk guardrails."
"The output isn't just a tool — it's a Quality Advisor. Healthcare frames administrative burden vs clinical quality as a zero-sum game. Solace Coding found a third way."
"A second set of expert eyes — running silently, surfacing what matters, so the clinician's first set can stay on the patient."
"After a heavy day of fighting with the technology, an evening of focused building didn't drain me — it restored me. Like exhale. Like solace."
Each tool is a multi-agent pipeline — specialist AI agents working in parallel, passing findings forward, so the clinician gets a synthesised, evidence-based answer in seconds.
8 specialist AI agents assess every dimension of a safe discharge — simultaneously. Data collection, medical history, SDOH barriers, clinical guidelines, LACE risk scoring, care planning, physician decision, and clinical note — all in ~16 seconds.
An 8-agent pipeline that manages the full discharge referral workflow — clinical triage, urgency scoring, referral letter drafting, specialist matching, patient communication, and follow-up coordination. Transforms a 45-minute admin task into a 20-second intelligent handoff.
An AI-enabled renal health screening and risk stratification tool for Nepali labor migrants in Gulf Cooperation Council countries — addressing CKDnT (heat-stress nephropathy) through pre-departure screening, in-country monitoring, and returnee assessment. Covers occupational heat-stress prevention, tele-nephrology, a national CKD clinical registry, and community awareness campaigns reaching workers, families, and primary care providers across Nepal's seven provinces and GCC worksites.
Paste a consultation transcript or summarise a patient encounter — the AI produces a structured SOAP note, auto-codes diagnoses, and generates a patient-facing summary in plain language. Built for the 45 minutes a day physicians spend on documentation.
An agent that continuously monitors population-level SDOH signals — food insecurity, housing instability, transportation barriers — and surfaces high-risk cohorts for proactive outreach before crises occur.
Unlike a single AI model that does everything, agentic pipelines assign each domain to a specialist agent — then orchestrate their findings into a unified clinical output.
Six specialist agents fire simultaneously — Data Collector, Medical History Analyst, SDOH Assessor, Guidelines Checker, Risk Analyst, Plan Synthesizer. Each reads the patient case and produces an expert analysis in its own domain. No waiting. No bottleneck.
Once all six analyses land, a dedicated Decision Agent reads every report holistically — just as a consultant physician would review specialist notes — and issues a structured GO / NO-GO recommendation with LACE score, primary rationale, and colour-coded considerations.
The final agent reads all upstream outputs and writes a complete, signed-ready clinical document. GO verdict → full Discharge Summary. NO-GO verdict → Discharge Assessment Note documenting barriers and the plan to overcome them. Print-ready. Copy-to-clipboard. Done.
The physician reviews the synthesised recommendation and clinical note — not writes it from scratch. The cognitive load shifts from documentation to clinical judgement. That's the Solace Coding promise: technology in service of calm.
Clinical decisions are multidimensional. A single AI cramming everything into one response misses the nuance. Specialist agents, working in parallel, get it right.
A dedicated SDOH agent ensures social determinants are assessed on every case — not when the clinician remembers to ask.
The LACE index is calculated by a dedicated Risk Analyst agent — objective, consistent, and evidence-based every time.
A Guidelines Checker agent cross-references AHA, IDSA, AAOS protocols — catching deviations before discharge.
The Clinical Note Writer produces a complete, structured clinical document — the physician reviews and signs, not writes.
Every Solace Agent tool is built against all five dimensions of the Quintuple Aim. Not as a checkbox. As a design constraint.
Click a command to explore the philosophy, tools, and clinical thinking behind Solace Agent for Healthcare.
Both tools are live, free, and need only your API key (Anthropic or OpenAI). No installation. No server. Open in browser and run.