Lead Full-Stack AI Engineer

Remote, full-time role for a Lead Full-Stack AI Engineer based in India. You will own the core platform of an AI-enabled drug-development product, working directly with the founder and a US-based Forward Deployed Engineer. A daily 3-hour overlap with US Pacific Time (typically Indian evenings) is required.
Lead Full-Stack AI EngineerEngagement Model: Full-Time / Employer of Record (EOR)Compensation: Competitive Market Rate (Commensurate with Full-Stack & Multi-Agent AI Depth)Location: India (Bengaluru, Hyderabad, or NCR preferred)Target Experience: 5+ years of full-stack software engineering with demonstrable AI/Multi-Agent system depth, including 2+ years building and deploying production LLM / Multi-Agent AI systems.
About INCEVO LLCINCEVO LLC is a biopharmaceutical consulting firm delivering end-to-end program leadership, functional expertise, and BD due diligence to help life sciences companies advance therapies from discovery through regulatory approval. INCEVO is pioneering a proprietary AI-enabled drug-development platform— integrating real-time intelligence, regulatory insight, and predictive analytics to accelerate decision-making and de-risk portfolio investments — positioning the firm at the forefront of next-generation biopharma strategy.
About the RoleWe are seeking a Lead Full-Stack AI Engineer to take complete operational ownership of our core platform codebase—from database schemas and asynchronous backend microservices to responsive, highly polished user interfaces. In this high-velocity, high-agency role, you don't just write basic API wrappers; you build resilient, high-throughput systems that turn complex AI capabilities into fast, reliable enterprise product. You will work directly with the founder and our Forward Deployed Engineer to iterate product features, optimize LLM inference speeds, and enforce rigorous engineering standards across our stack.
Why This Role Matters: As our Lead Full-Stack AI Engineer, you will own the technical engine powering our platform and drive it across the finish line into a polished, hardened, enterprise-ready product that performs flawlessly in live client programs. The domain problem is genuinely novel: orchestrating drug development with AI agents is an unsolved enterprise challenge, and you will build the UI components, backend microservices, and system reliability required to make it seamless for global pharma clients. If you want to architect production-grade agentic AI in one of the most important industries in the world, this is that role.
What You’ll Do· Drive Enterprise Production Polish & New Feature Surfaces: Refine, refactor, and expand our core web application (React, Next.js, TypeScript, Tailwind) to elevate it from a functional platform into a high-performance, enterprise-grade product. Own the addition of new UI components, micro-interactions, and backend workflows while maintaining clean, scalable frontend architecture. Translate client program needs into intuitive UI components with minimal friction. Incorporate architectural insights surfaced by our Forward Deployed Engineer from live client deployments to ensure new product surfaces reflect real enterprise client workflow needs.· Architect Low-Latency Agentic Backends: Build and maintain asynchronous REST and WebSocket microservices (FastAPI / Node.js) handling multi-agent state persistence, tool calling, inter-agent orchestration (LangGraph), streaming tokens responses, and parallel data/document ingestion pipelines. Ensure every API endpoint is fast, reliable, and self-documenting. Performance and reliability under concurrent client load are non-negotiable.· Optimize AI & RAG Pipelines: Implement, benchmark, and maintain production vector retrieval pipelines, hybrid search (BM25 + dense vector), evaluation harnesses (Ragas, TruLens), and multi-tier caching structures (Redis / semantic caching). Own execution repeatability and token efficiency across all LLM workflows.· Ensure System Reliability & Observability: Write robust test suites (PyTest, Playwright), implement structured data validation (Pydantic), and build fallbacks and circuit-breakers for when downstream LLM APIs experience latency or outages. Proactively surface system bottlenecks before users encounter them — and translate reliability findings into concrete architectural improvements that make the platform progressively more resilient across every client deployment cycle.· Maintain Engineering Hygiene: Strong Git discipline (atomic commits, clean rebase/PR workflows), structured pull requests, self-documenting code, and end-to-end CI/CD ownership (GitHub Actions). Every commit carries clear context and intentional changes.· Collaborate Asynchronously: Operate autonomously with clear written documentation and maintain a mandatory 3-hour daily overlap with US Pacific Time for daily standup, sprint planning, and real-time collaboration. Work in close partnership with our US-based Forward Deployed Engineer — translating field-validated architectural requirements from live client deployments into platform improvements rapidly and precisely. Own the full handoff from architectural specification to live production code.
You Might Thrive in This Role If You Have:· Startup Velocity & High Agency: You care deeply about shipping clean, self-documenting code fast and iterating rapidly based on real user feedback. You flag blockers early and propose architectural solutions, not just problems.· 5+ Years of Full-Stack Engineering Depth: 5+ years of full-stack software engineering with demonstrable AI/Multi-Agent system depth, including 2+ years specifically spent building, deploying, and architecturally evolving production-grade LLM or Multi-Agent AI systems at enterprise scale. Strong proficiency across TypeScript, React/Next.js, Python (FastAPI/Pydantic), PostgreSQL, and Redis.· Multi-Agent Systems & AI Expertise: Hands-on production experience with modern agentic frameworks (LangGraph, AutoGen, LlamaIndex workflows), structured JSON outputs, vector databases (Qdrant, Azure AI Search, pgvector), and API integrations (OpenAI, Anthropic Claude, NVIDIA NIM). You understand how LLM-based systems fail in production differently from traditional CRUD apps — and you make deliberate architectural decisions that prevent, detect, and recover from those failure modes at scale.· Async & Performance Optimization Mastery: Deep working knowledge of Python asyncio, streaming token consumption, memory management, time-to-first-token (TTFT) reduction, and API rate-limit/retry handling. You know when to reach for a WebSocket versus a polling endpoint and why it matters for perceived performance.· Azure Stack Familiarity: Practical experience with Azure Container Apps, Azure PostgreSQL, Azure Blob Storage, and Azure AI Search. Familiarity with Azure AD SSO and Key Vault is a strong advantage. You treat client program data with the same care a regulated pharmaceutical company would.· Engineering Hygiene & IP Security: Strong Git workflow discipline (atomic commits, clean PR reviews, rebase hygiene), full CI/CD ownership via GitHub Actions, and clear written technical communication in English that non-technical stakeholders can act on. You treat INCEVO's codebase, agent architecture, and client program data as proprietary assets — maintaining strict IP security standards at every stage of development, deployment, and offboarding.· Cross-Functional Engineering Collaboration: You work effectively with a US-based Forward Deployed Engineer who owns client-facing deployments and brings real-world architectural requirements from the field. You translate those requirements into platform improvements rapidly and precisely — closing the loop between client reality and core engineering execution.· Biopharma or Life Sciences Context (Significant Advantage): Prior experience building software for pharmaceutical, clinical research, or regulated healthcare environments — including familiarity with clinical data schemas (ClinicalTrials.gov, CDISC), regulatory submission workflows, or GxP system validation concepts — is a meaningful differentiator. The founding team provides full domain context; what we need from you is the technical stack mastery and execution speed.
Equal Opportunity & Confidentiality· Equal Opportunity Employer: INCEVO LLC is an equal opportunity employer. We evaluate all qualified candidates regardless of race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, or other legally protected characteristics. We believe that building AI systems for high-stakes, complex domains requires diverse perspectives, deep intellectual curiosity, and rigorous engineering standards.· Work Authorization & Intellectual Property: All offers and engagements are contingent upon the successful execution of INCEVO LLC’s standard Non-Disclosure Agreement (NDA) and Intellectual Property (IP) Assignment Agreement prior to commencement. For international engagements, employment is administered in full compliance with local labor laws via our authorized Employer of Record (EOR) platform.

Skills
Python, JavaScript, TypeScript, React, Node.js, Next.js, FastAPI

Experience
2+ yrs

Work arrangement
Yes