21 Nov
|
HCLTech
|
Hyderabad
Job Summary
We are seeking a hands-on AI/ML Engineer to design, build, evaluate, and productionize enterprise-grade Agentic AI solutions. The role requires strong Python engineering, practical experience with large language models and agent frameworks, and the ability to create secure, observable, and reliable autonomous workflows that integrate with enterprise platforms and data sources.
Key Responsibilities
- Design and implement autonomous AI agents for multi-step reasoning, planning, goal decomposition, task orchestration, and dynamic decision-making.
- Build single-agent, hierarchical, and multi-agent solutions using patterns such as ReAct, Plan-and-Execute, reflexive agents, and tool-augmented agents.
- Develop custom orchestration layers and integrate agents with internal APIs, databases, SaaS platforms, CRM/ERP systems, and enterprise services.
- Implement state, short-term memory, long-term memory, context management, and Retrieval-Augmented Generation (RAG) capabilities.
- Engineer robust tool execution with validation, timeouts, retries, privilege controls, sandboxing, failure detection, and graceful recovery.
- Create reusable prompt, planning, critique, reflection, and self-correction patterns for reliable instruction following and tool use.
- Deploy AI agents to production and establish tracing, logging, prompt/model versioning, monitoring, evaluation, and CI/CD practices.
- Design automated evaluation frameworks covering task success, cost, latency, reliability, safety, hallucination detection, and regression testing.
- Apply Responsible AI and security controls, including prompt-injection mitigation, data isolation, explainability, bias awareness, and human-in-the-loop escalation.
- Collaborate with architects, product owners, data engineers, security teams, and business stakeholders to translate use cases into scalable solutions.
Skill Requirements
- Programming and systems: Expert-level Python; asynchronous programming, concurrency, task scheduling,
API development, and software engineering best practices.
- Agentic AI: Hands-on development of autonomous agents, tool/function calling, planning and re-planning, stateful/stateless design, memory management, and multi-agent orchestration.
- LLMs and prompting: Experience with Azure OpenAI, OpenAI, Anthropic, or open-source models; zero-shot/few-shot prompting; reasoning, planning, reflection, critique, and model-selection trade-offs.
- Frameworks: Practical experience with LangGraph/LangChain, Semantic Kernel, AutoGen, CrewAI, or equivalent agent frameworks; ability to work beyond framework abstractions.
- RAG and knowledge: Vector databases and search platforms such as FAISS, Pinecone, or Azure AI Search; embeddings, chunking, retrieval, reranking, context compression, and hybrid retrieval.
- MLOps and AgentOps: Production deployment, tracing, observability, model/prompt versioning, CI/CD, Docker, Kubernetes, and serverless deployment on Azure or AWS.
- Evaluation: Offline test harnesses, simulation, A/B testing, task-quality metrics, safety testing, hallucination detection, cost analysis, and latency/reliability measurement.
- Data and integration: REST APIs, web services, SQL/NoSQL databases, enterprise integration patterns, authentication, secrets management, and working knowledge of event-driven systems.
- Cloud: Strong hands-on experience with Azure, AWS, or GCP; managed AI services, identity and access management, security, and LLM cost optimization.
Must Have Skill
Large Language Models (LLMs),Agentic AI,Generative Artificial Intelligence (Gen AI)
Good to Have Skill
Other Requirements
Preferred / Nice-to-Have Skills
- Fine-tuning techniques and adapters such as LoRA.
- Knowledge graph-augmented generation, GraphRAG, or hybrid memory architectures.
- Multi-agent collaboration, negotiation, and supervisor-worker patterns.
- Human-AI collaboration patterns, enterprise copilots, and autonomous workflow development.
- Reinforcement learning or policy optimization for agent behavior.
- Experience in regulated enterprise environments with strong privacy, security, auditability, and governance requirements.
Qualifications
- Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Science, Engineering, or a related discipline.
- 8-10 years of software engineering, machine learning, or AI solution development experience, including hands-on Generative AI or Agentic AI delivery.
- Demonstrated ability to take AI solutions from proof of concept through production deployment and operational support.
- Strong analytical, troubleshooting, communication, and stakeholder-collaboration skills.
Expected Outcomes
- Production-ready agents that execute enterprise workflows safely and reliably.
- Measurable improvements in task success, automation, response quality, cost, and turnaround time.
- Reusable agent components, evaluation assets, technical documentation, and operational runbooks.
- Compliant solutions with explicit observability, controlled tool access, auditability, and human oversight.
Key Screening Indicators
- Has personally built and deployed an agent that uses multiple tools or enterprise APIs.
- Can explain state, memory, planning, retries, guardrails, and evaluation using a real implementation example.
- Understands the trade-offs between deterministic workflows and autonomous agent behavior.
- Can demonstrate production considerations such as security, monitoring, latency, cost, and failure recovery.
📌 Senior Technical Lead (Hyderabad)
🏢 HCLTech
📍 Hyderabad