Introduction
A career in IBM Consulting embraces long-term relationships and close collaboration with clients across the globe.
You’ll work with visionaries across multiple industries to enhance the hybrid and AI journey for some of the world's most innovative and valuable companies. Your ability to accelerate impact and effect meaningful change for your clients is supported by our strategic partner ecosystem and our comprehensive technology platforms, including IBM Software and Red Hat.
Curiosity and a constant quest for knowledge form the foundation of success in IBM Consulting. In your role, you’ll be encouraged to challenge the norm, explore ideas beyond your immediate responsibilities, and devise creative solutions that result in groundbreaking impact for a wide network of clients. Our culture of evolution and empathy focuses on long-term career growth and development opportunities in an environment that celebrates your unique skills and experiences.
In this role, you'll work in one of our IBM Consulting Client Innovation Centers (Delivery Centers), which deliver deep technical and industry expertise to a diverse range of public and private sector clients around the world. Our delivery centers provide clients with locally based skills and technical expertise to foster innovation and promote the adoption of new technology.
Your Role and Responsibilities
We are seeking a Lead AutoGen Engineer with extensive experience in Autogen. You should have progressed beyond merely building RAG pipelines to designing systems that ensure agentic AI operates reliably in regulated, high-stakes environments for both client-facing and internal platforms.
This is a hands-on technical leadership role ideal for someone who has grappled with the challenges posed by complex regulatory language, constructed agentic workflows from scratch, and approaches LLM outputs much like a systems engineer considers correctness guarantees. You will be responsible for designing and building production-grade agentic systems—not merely as a research exercise but as engineering deliverables that function reliably in compliance-driven environments. The problems you will tackle are not suited for off-the-shelf solutions. You will be expected to make architectural decisions, iterate when initial designs prove inadequate, and communicate those decisions clearly to both technical teams and enterprise stakeholders.
What You Will Do
- Design and deliver RAG systems for complex, document-heavy domains, defining ingestion strategies, retrieval architectures, and generation-side optimization appropriate to the content type and use case.
- Architect multi-agent orchestration systems that coordinate specialized agents using autogen and other frameworks across parallel and sequential workflows, incorporating structured quality evaluation and targeted correction loops in the pipeline.
- Build LLM routing and evaluation infrastructure—selecting models per task, optimizing generation parameters, and ensuring full auditability of outputs for regulated delivery environments.
- Lead end-to-end technical delivery: from problem framing and system design through implementation, quality assurance, and effective stakeholder communication.
- Work across regulated industries, including life sciences, insurance, and financial services, grasping what compliance, traceability, and domain accuracy entail at the implementation level.
- Evaluate and apply fine-tuning approaches when retrieval-based solutions fall short, with the discernment to know when each method is appropriate.
- Mentor engineers and establish technical standards for the development of agentic AI within the practice.
- Collaborate with client stakeholders to translate complex domain requirements into actionable system behaviors.
What We Offer
- Individual career path planning.
- Access to countless training and certification offerings.
- A culture of knowledge sharing.
- An environment that encourages you to contribute your ideas actively.
- The opportunity to work on projects within one of the largest IT companies.
Required Technical and Professional Expertise
- Proven experience (8+ years) in building and iterating on production-grade agentic systems across the full stack—ingestion, retrieval, orchestration, generation, and quality assurance.
- Deep understanding of statistics, probability, data science, machine learning, and transformer architectures.
- Expertise in multi-agent frameworks—designing orchestration patterns, parallel agent architectures, feedback loops, and structured output pipelines.
- Comprehensive knowledge of data ingestion and RAG pipeline designs across diverse content types, retrieval strategies, and optimization techniques.
- Extensive experience delivering agentic systems end-to-end within production pipelines in regulated industries.
- Experience integrating multiple LLM providers with routing logic, audit logging, and parameter optimization for enterprise and regulated contexts.
- Advanced skills in prompt and context engineering.
- Integration of agentic solutions with cloud platforms such as Azure OpenAI, IBM Watsonx, and AWS, ensuring governance, safety, and monitoring.
- Proficiency in Python and relevant engineering tools; experience with cloud platforms, including AWS and GCP.
- Successful track record in delivering agentic systems with measurable operational impacts for enterprise clients in regulated industries.
- Advanced database skills across SQL (MySQL, SQL Server), NoSQL (MongoDB, Neo4j), and VectorDBs (CosmosDB, Azure AI Search, FAISS).
- Able to lead small technical teams, set direction, and maintain delivery quality across complex engagements.
- Strong communication skills—capable of presenting architectural decisions and system behaviors to non-technical audiences, including clinical, regulatory, and business stakeholders.
- Language proficiency in Hindi and English: professional-level required for coordination with other teams embedded in the engagement.
Preferred Technical and Professional Experience
- Experience with clinical or regulatory document workflows—pharmaceutical submissions, compliance reporting, or structured document generation at scale.
- Experience working with IBM's agentic asset portfolio.
- Familiarity with Model Context Protocol (MCP) and emerging agentic interoperability standards, skills, and tools.
- Multi-agent system design demonstrating a breadth of agentic architecture application across multiple domains.
Apply online using the form below. Only applications matching the job profile will be considered.