Lead Engineer - Agentic AI

Datazymes Analytics Pvt Ltd; Bengaluru; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

DataZymes Analytics Pvt. Ltd. is seeking a builder-leader to architect and ship multi-agent systems across pharma data pipelines and regulatory intelligence workflows.

You will write production-grade code, own deployments, and lead a small team delivering agentic AI solutions that meet strict governance and validation standards. Ideal candidates bring deep agentic frameworks experience, RAG and knowledge graphs, and strong collaboration with client teams.

Qualifications

• Strong background in building scalable AI systems with production-grade quality.
• Experience designing agentic frameworks and RAG pipelines for regulated domains.
• Ability to lead a small team and deliver with client focus.

Responsibilities

• Architect and ship autonomous agentic systems across pharma data pipelines.
• Define guardrails, monitoring, and human-in-the-loop escalation points.
• Lead a team of AI engineers and interface with client stakeholders.

Skills

Python

Agentic AI

LangChain

Team Leadership

Education

Bachelor's or Master's in CS/SE

Pharma data domain knowledge

Tools

LangGraph

OpenAI API

Google Vertex AI

Veeva / IQVIA integrations

Job description

DataZymes Analytics Pvt. Ltd. | Full time

We empower the Pharma industry with our innovative products.

The idea of DataZymes germinated with the realization that Pharma Commercial teams had few alternatives to the antiquated and inefficient solutions offered by traditional consulting and technology companies. Already a laggard in analytical maturity, the Pharma industry had been facing challenges to adapt to a Big Data world.

We saw that the products offered by technology companies were too rigid and generic to handle novel problems. The custom solutions offered by consulting organizations took too long to deploy and required many services to maintain and improve.

We felt the need for a different approach to finding solutions and we knew it would take a different kind of company to build it. That's why DataZymes.

We're focused on creating the world's best user experience for working with data, one that empowers people to ask and answer complex questions without requiring them to master querying languages, statistical modeling, or the command line. To achieve this, we are building platforms for integrating, managing, and securing data on top of which we layer applications for fully interactive machine-driven, human-assisted analysis.

Job Description

We are looking for passionate and driven professionals to join DataZymes, a next-generation analytics and data science company founded in 2016. At DataZymes, we focus on driving technology-led innovation and helping clients maximize the value of their data and analytics investments through cutting-edge platforms and consulting expertise. If you are excited about working on impactful solutions in the healthcare analytics space and want to be part of a high-performance, fast-growing team, we’d love to hear from you.

We are a data and analytics servicesfirm purpose-built for the pharmaceutical and life sciences industry. Ourclients span

This role is not a strategy position. It is not a research position. It is a builder-leader role. You will architect and shipmulti-agent systems that operate autonomously across pharma data pipelines,regulatory intelligence workflows, and cross-functional analytics use cases.You will write code, own production deployments, and lead a small team doingthe same.

The ideal candidate has deep technicalfluency in agentic frameworks, understands the compliance,data governance, and validation expectations of the pharma industry, and can translate both into workingsystems, not slide decks.

Requirements

• Design and build end-to-end agentic systemscombining LLMs, multi-agent orchestration, enterprise data pipelines, andpharma-specific business logic. Ship production-grade systems, not prototypes.
• Select and implement the right orchestrationapproach across no-code, low-code, and pro-code patterns based on use casecomplexity and client readiness.
• Architectretrieval and knowledge services (RAG, knowledge graphs) over structured andunstructured pharma data: Rx, claims, engagement, clinical trial data, RWEdatasets, label text, and scientific literature. Includes RAG pipelines,knowledge graphs for entity-relationship modeling (HCP, drug, indication, trialnetworks), hybrid search, and retrieval evaluation frameworks.
• Build observability, monitoring, andevaluation frameworks to track agent behavior in production. Define guardrails,failure modes, and human-in-the-loop escalation points.
• Integrate with upstream pharma data platforms(IQVIA, Symphony, Komodo, Veeva) and downstream delivery surfaces via APIs andworkflow hooks.

PHARMA DOMAIN APPLICATION

• Translatecommercial analytics, medical affairs, and clinical operations workflows intoagentic automation opportunities. Target high-volume, high-complexity,logic-intensive processes first.
• Build agentsthat operate over 21 CFR Part 11-aware environments. Understand whatauditability, validation, and traceability mean for autonomous systems in aregulated context.
• Developintelligent document processing pipelines for clinical study reports, druglabels, HEOR submissions, and payer dossiers.
• Applyagentic AI to KOL identification and mapping, literature synthesis, competitiveintelligence, and signal detection workflows.

LEADERSHIP & CLIENT DELIVERY

• Lead a team of AI engineers and MLpractitioners. Set technical direction, review architecture decisions, andmaintain a high bar for production quality.
• Partner with client-facing teams to scopeagentic AI engagements: define the use case, design the solution architecture,and own delivery accountability.
• Communicate complex agent system behavior tonon-technical pharma stakeholders. Bridge the gap between what agents do andwhat the business needs to trust.
• Champion AI governance practices aligned withindustry standards: documented agent decision logic, bias audits, andtraceability to source data.
• Build internal capability by mentoring teammembers and establishing the firm's agentic AI playbook as a reusable asset.

WhatYou Bring

TECHNICAL DEPTH (REQUIRED)

• 8+ years in software or MLengineering; 3+ years with production LLM or agentic AI systems.
• Hands-on proficiency with agenticframeworks: LangGraph, LangChain, AutoGen, CrewAI, or equivalent. Model ContextProtocol (MCP) familiarity strongly preferred.
• Direct SDK experience: Anthropic(Agents SDK, tool use, Claude API), OpenAI (Assistants API, function calling),Google (Vertex AI Agent Builder, Gemini API). Model Context Protocol (MCP)strongly preferred.
• Python fluency. Ability to build,test, and deploy production code, not just notebooks.
• Strong RAG architecture skills:chunking strategies, embedding models, vector stores, knowledge graphs forentity-relationship modeling (drug-indication-HCP-trial), hybrid search,retrieval evaluation.
• Observability tooling for AIsystems: logging agent traces, eval frameworks, cost management, driftdetection.

PHARMA / LIFE SCIENCES DOMAIN (REQUIRED)

• Working knowledge of pharma commercial dataecosystems: Rx/claims data, NPI-level analytics, market access, brandperformance
• Familiarity with regulated data environments: GxP,21 CFR Part 11, HIPAA-compliant data handling, audit trail requirements
• Exposure to at least two of: medical affairsanalytics, real-world evidence, clinical operations data, or HEOR/market accessworkflows
• Comfort reading and reasoning over scientific andregulatory documents: labels, clinical study reports, AMCP dossiers, payerbriefs

LEADERSHIP & COMMUNICATION (REQUIRED)

• 5+ years leading technical teams or deliveryworkstreams, including mentoring engineers and managing project scope andtimelines
• Track record of shipping production AI solutionswith measurable business impact, not just proof-of-concepts
• Comfortable in executive-level conversations:scoping engagements, presenting architecture trade-offs, and aligning ongovernance expectations
• Strong written communication. You can write acrisp technical spec and a clear client-facing proposal without switching tools

GOOD TO HAVE

• Experience with Veeva Vault, Medidata, or IQVIAplatform integrations
• Knowledge of reinforcement learning from humanfeedback (RLHF) and fine-tuning workflows
• Familiarity with EU AI Act and emerging FDAguidance on AI/ML in clinical and regulatory contexts