Senior Applied AI Engineer

Senior Applied AI Engineer

About the role

Senior Applied AI Engineer @ August AI Full Time • Bangalore, India August is building autonomous clinical AI for the world. The product today shows up as a 24/7 health companion: people use August to understand symptoms, lab reports, prescriptions, medications, and the confusing space between doctor visits. But the company we are building is bigger than a companion. We are building the clinical reasoning, memory, safety, and workflow infrastructure for AI systems that can take on increasingly complex clinical work — accurately, empathetically, and with the right guardrails. This is not a generic chatbot role. You will work on AI systems that reason across patient history, symptoms, lab data, clinical guidelines, medical literature, and real-world user conversations. You will build agentic clinical workflows, evaluation systems, safety layers, escalation mechanisms, and production infrastructure that move August closer to clinical-grade autonomy. We are looking for a senior engineer who can take ambiguous, high-stakes problems and turn them into reliable product systems. You should be comfortable moving across backend engineering, applied AI, LLM systems, data pipelines, evals, and product judgment. The ideal person has strong engineering taste, high ownership, and a deep desire to build AI that materially improves healthcare access.

What you’ll build

• Autonomous clinical reasoning systems Build AI workflows that can reason through symptoms, patient context, lab reports, medications, risk factors, and possible next steps. This includes differential reasoning, triage, care navigation, escalation logic, and clinical decision-support workflows.
• Agentic clinical architecture Design multi-agent and verifier systems where specialized models handle different parts of the clinical task: history-taking, differential diagnosis, guideline retrieval, medication reasoning, lab interpretation, safety review, uncertainty detection, and escalation.
• Clinical knowledge and RAG infrastructure Build retrieval systems over trusted medical sources, clinical protocols, guidelines, user health history, prescriptions, reports, and structured medical content. The goal is to make August’s reasoning grounded, current, explainable, and auditable.
• Evaluation and safety systems Create eval suites, simulated-patient tests, regression benchmarks, red-team cases, clinical safety metrics, and monitoring dashboards. You will help define how August measures diagnostic quality, escalation accuracy, empathy, hallucination risk, refusal quality, and clinical reliability.
• Real-world health product experiences Ship AI features across WhatsApp, mobile, and web experiences. August should feel simple to users, even when the system underneath is reasoning across complex medical context.
• Production AI infrastructure Build low-latency, cost-efficient, observable systems that can scale to millions of users and handle high-stakes conversations reliably.

Responsibilities

• Own AI product systems end to end: problem definition, architecture, implementation, evaluation, launch, monitoring, and iteration.

• Build backend services, data pipelines, and model orchestration systems for clinical AI workflows.

• Develop agentic reasoning systems, RAG pipelines, model-routing logic, prompt/context systems, and verifier layers.

• Create robust evaluation frameworks for clinical accuracy, safety, empathy, escalation, and product quality.

• Work with doctors, medical reviewers, product, design, and engineering to translate clinical judgment into reliable software systems.

• Build safety mechanisms for high-risk situations, including uncertainty detection, emergency escalation, human review, and fail-safe behavior.

• Improve August’s ability to reason from messy real-world inputs: incomplete histories, noisy symptoms, PDFs, lab reports, prescriptions, and longitudinal user context.

• Debug production issues across the application, data, model, and infrastructure layers.

• Help define August’s engineering standards for clinical AI reliability, privacy, testing, observability, and speed of execution.

• You might be a good fit if you have

• 5+ years of experience building and shipping production software.

• Strong backend engineering experience, especially with Python.

• Experience building production LLM, ML, or data-intensive systems; not just prototypes.

• Strong understanding of APIs, distributed systems, testing, monitoring, and production reliability.

• Experience with some combination of RAG, prompt engineering, context engineering, agents, fine-tuning, model routing, evals, observability, or model deployment.

• High product judgment: you care whether the system actually helps users, not just whether the model appears impressive.

• Comfort working in ambiguous, high-stakes problem spaces where the answer is not obvious.

• Strong written and verbal communication. You should be able to explain technical tradeoffs to engineers, doctors, and product teams.

• Bias toward ownership. You do not wait for perfect specs; you clarify the goal, make tradeoffs, and ship.

• Nice to have

• Experience in healthcare, clinical AI, medical NLP, health data, or regulated environments.

• Experience with clinical guidelines, medical literature retrieval, medication reasoning, lab-report interpretation, triage, or differential diagnosis systems.

• Experience with FHIR, HL7, EHR integrations, QHINs, claims data, or healthcare data pipelines.

• Experience building safety-critical AI systems with human review, audit trails, escalation protocols, and measurable quality gates.

• Experience with PyTorch, TensorFlow, FastAPI, LangGraph, LangChain, LlamaIndex, vector databases, eval frameworks, or LLM observability tools.

• Experience building multilingual, mobile-first, WhatsApp-first, or voice-based AI products.

• Startup or founding-engineer experience.

• How we work

• High ownership

• We are a small team building in uncharted territory. You will own outcomes, not tickets. When something matters, you will take it from vague idea to safe production launch.

• Clinical safety first

• We move fast, but not recklessly. In healthcare, quality, escalation, privacy, and user trust are product features.

• Direct feedback

• We give clear, honest feedback and expect low-ego collaboration. The standard is high because the work matters.

• Obsessive craft

• We care about the details: the wording of a response, the accuracy of an escalation, the latency of a workflow, the design of an eval, and the reliability of a system under stress.

• Pragmatic AI

• We do not build AI demos. We build systems that work in production, with measurable quality, clear failure modes, and fast iteration loops.

• Why August

• Healthcare systems around the world are overloaded. Millions of people are left alone with symptoms, reports, prescriptions, and uncertainty at exactly the moments when they need clarity most.

• August is building AI that can be available in those moments: accurate, empathetic, clinically grounded, and increasingly capable of doing real clinical work. We already operate at meaningful scale, and we are just getting started.

• You will join early enough to shape the architecture, culture, and safety standards of the company, but late enough that your work will reach real users quickly.

• Must-have skills

• Pythonbackend engineeringLLM systems

• Good-to-have skills

• RAGprompt engineeringcontext engineering