Role Purpose
Engineering Lead owns the full engineering lifecycle of MaxisAI's clinical AI platform, from architecture and development to CI/CD, quality, and technical debt management, and delivers scalable, secure, enterprise-grade software. This is a hands-on player-coach role: roughly 60% hands-on architecture and coding, 40% team management.
You will personally design, code, and ship the agentic AI platform on AWS Bedrock, LangChain, and LangGraph, along with its microservices and PostgreSQL/MongoDB data layer, and run an AI-native development lifecycle end to end.
Day to day, you will write the agent code yourself. That covers Lang Smith, LangGraph multi-agent workflows, tool and function calling, agent memory, RAG pipelines on Bedrock Knowledge Bases, Guardrails, and LangSmith evals. You will debug agents in production from their traces and tune prompts and models for accuracy, cost, and latency. You will also turn working prototypes into production-grade, GxP-compliant services the team can extend.
Key Responsibilities
- Own architecture, engineering delivery, and technical quality across all MaxisAI product components.
- Lead, mentor, and grow a high-performing engineering team of frontend, backend, and platform engineers.
- Drive adoption of AI-assisted development tools (Claude Code, GitHub Copilot, Cursor) across the engineering team.
- Own the technical roadmap and engineering capacity planning, aligned to product and business priorities.
- Work with the Product Leader to refine requirements, estimate effort, and commit to delivery timelines.
- Oversee platform reliability, performance, and availability across multi-tenant and single-tenant deployments.
- Architect and build multi-agent systems: orchestration, planning, tool/function calling, memory, and human-in-the-loop workflows.
- LangChain, LangGraph, LangSmith
- AWS Bedrock: Agents, AgentCore (runtime, memory, gateway, identity), with pgvector, Aurora PostgreSQL, or OpenSearch Serverless.
- Design and build microservices using domain-driven design, an API gateway, and event-driven patterns (SQS/SNS/EventBridge/Kafka).
- Deploy on AWS using EKS/ECS, Lambda, and Step Functions, with IaC in Terraform or CDK.
- Extensive Hands-On Exp – AIDLC:
- AI-assisted user stories
- Agentic coding with Claude Code, Cursor, and Copilo
- AI first-pass review for quality, security, and GxP checks
- AI quality gates, eval-gated deployment
- AI-driven incident triage, log and trace analysis, and self-healing runbooks
- Manage vendor and third-party technology relationships relevant to engineering.
- Ensure adherence to GxP software validation, data security, and compliance requirements.
- Conduct architecture reviews, technology evaluations, and build-vs-buy decisions.
- Partner with the Delivery Excellence to align engineering velocity with SDLC standards.
- Run model selection and evals across Claude, Llama, and Titan using Bedrock Model Evaluation.
- Optimise cost and latency with prompt caching, provisioned throughput, and batch inference.
- Establish LLMOps: prompt versioning, eval frameworks, and hallucination controls.
- Ensure AI outputs are traceable, auditable, and explainable for clinical and GxP use.
- Documentation & Validation: AI-assisted docs and GxP artefacts (IQ/OQ/PQ, traceability matrices).
- Define guardrails, human-review checkpoints, and audit trails for all AI-generated code and artefacts.
- Use Claude Code, Cursor, and Copilot daily to model AI-native development.
Required Experience
9 – 10 years in software engineering, including at least 4 years in an engineering lead role. 2-3 Yrs on Agentic AI experience is Mandatory preferably in Pharma/LifeSciences/CRO’s/SiteNetworks/Healthcare Industry.
- Experience integrating or building AI/ML-powered products or workflows.
- Has personally architected and shipped production agentic AI or multi-agent systems.
- Hands-on production experience with LangChain, LangGraph, and LangSmith.
- Hands-on production experience with AWS Bedrock (Agents/AgentCore, Knowledge Bases, Guardrails).
- Strong microservices architecture experience: distributed, event-driven systems at scale.
- Proven end-to-end experience running an AI-native or AI-augmented ADLC, with measurable outcomes.
- Experience with GxP software validation or FDA-regulated system delivery.
- Good hands-on experience in AWS Bedrock, LLM orchestration, or agentic AI architectures.
- AWS certification: Solutions Architect Professional or Machine Learning Specialty.
- Experience with Strands Agents, MCP, and vector databases.
- Validation: AI-assisted docs and GxP artefacts (IQ/OQ/PQ, traceability matrices).
- Define guardrails, human-review checkpoints, and audit trails for all AI-generated code and artefacts.
- Agentic system design round.
- Hands-on background in backend (Node.js, Python), frontend (React), and cloud-native architectures (AWS preferred).