Lad AI Engineer

Netscribes; Bengaluru; On-site; INR 1,200,000 - 1,800,000; Full time

Job summary

Netscribes in Bengaluru seeks an experienced ML/AI engineer to lead end-to-end ML foundation, lifecycle, and MLOps. You will design data prep, training, deployment, and monitoring pipelines, manage model and data drift, and enforce engineering standards across teams.

Responsibilities span agentic AI engineering, GenAI foundations, RAG, evaluation, and governance, with leadership responsibilities and client-facing delivery. This role operates within a CoE environment on site in Bengaluru.

Qualifications

• Experience in end-to-end ML pipeline design covering data prep, training, evaluation, deployment and monitoring.

Responsibilities

• Own end-to-end ML pipeline design covering data preparation, feature stores, training, evaluation, deployment, and monitoring.
• Instrument and manage model drift, data drift, concept drift with automated alerting and retraining workflows.
• Define and enforce ML engineering standards across the team: reproducibility, experiment tracking, code quality.
• Architect and build multi-agent systems for complex business workflows.
• Design agent communication patterns, tool registries, memory management, and failure recovery.
• Select, fine-tune, and serve foundation models; manage prompt-to-model drift and evaluation.
• Design retrieval systems and configure vector databases for RAG and knowledge graphs.
• Establish evaluation frameworks and lead technical delivery and governance discussions.

Skills

Python

ML Engineering

MLOps

LangGraph

AutoGen

CrewAI

Databricks

SageMaker

Vector DBs

RAG

LLM Evaluation

Tools

Databricks

AWS Bedrock

SageMaker

Azure ML

Pinecone

Weaviate

Qdrant

Milvus

Job description

About Netscribes Netscribes is a global leader in data, insights, and digital solutions, helping the worlds largest organizations accelerate growth and innovation. As a growth catalyst, we empower sales, marketing, product development, and strategy through a unique blend of domain expertise and technological capabilities. Our end-to-end solutions span data engineering, advanced analytics, AI, and intelligent automation built to scale and adapt to dynamic business environments. We partner with clients across the implementation journey, aligning with their ecosystem to deliver actionable intelligence, operational efficiency, and competitive advantage. For more information, visit www.netscribes.com.

LoB Overview Netscribes delivers integrated solutions across content, process, and technology to help global businesses scale efficiently and stay competitive in fast-moving markets. Our services are structured across three key areas:

Content Solutions We specialize in building, managing, and optimizing large-scale content ecosystems for global brands. Our expertise spans product data enrichment, taxonomy development, catalog and digital shelf management, content operations, and digital asset optimization. These solutions help improve product discoverability, streamline content workflows, and enhance customer experience across digital platforms.

Process Solutions Our research and insights-driven services support core business processes across marketing, product, customer experience, and operations. This includes market and competitive research, campaign analysis, customer intelligence, and content performance tracking. Through Information Management Services (IMS) and managed service models, we embed scalable, high-quality support directly into client workflows, ensuring agility, consistency, and speed.

Technology Solutions Complementing our content and process offerings, we bring advanced technology capabilities through data engineering, data analytics, and AI. From building unified data platforms and real-time dashboards to deploying intelligent automation and machine learning models, we help organizations modernize operations and unlock data-driven decisions at scale.

Profile Description

LOB: Support | CoE

• Python Production-grade; clean, modular, reviewable code; not notebook-first engineering Classical ML, Model Lifecycle & MLOps End-to-end ownership across the full stack: feature engineering, model training, evaluation, deployment, versioning, monitoring, drift detection (data, concept, prediction), retraining triggers, and CI/CD for ML; MLflow or equivalent; this should be lived delivery experience must be able to articulate what broke in production and why, not just what trained well.
• Agentic AI Engineering Design and build of multi-agent systems; tool use, agent orchestration, memory management (short and long-term), state handling; LangGraph, AutoGen, CrewAI or equivalent; must understand agent failure modes and recovery patterns, not just happy-path flows.
• Execution Platforms Hands-on experience across at least two of: Databricks (Mosaic AI, Agent Bricks, MLflow), AWS (Bedrock, SageMaker), Azure (AI Foundry, Azure OpenAI, Azure ML, Azure AI Services); platform-aware and capable of operating across environments without being locked to one.
• Foundation Model Engineering Fine-tuning (full, LoRA, QLoRA), structured prompt engineering, prompt-to-model drift management, evaluating output quality across model versions.
• RAG Architecture Strong fundamentals non-negotiable: document ingestion, chunking strategy, embedding model selection, vector store integration, re-ranking, retrieval evaluation; Graph RAG and knowledge graph-based retrieval are expected areas of growth prior exposure preferred, willingness and ability to lead implementation required.
• Vector Databases Hands-on experience with at least one: Pinecone, Weaviate, Qdrant, or Milvus; must understand trade-offs between options and be able to make selection decisions based on use case requirements.
• LLM Evaluation & Testing Structured evaluation using frameworks such as RAGAS, TruLens, DeepEval or equivalent; golden set benchmarking, LLM-as-judge, relevance scoring, hallucination detection; must be able to define and implement what good looks like quantitatively, not just qualitatively.
• AI Governance Baseline At this level of seniority, expected to speak to model observability, audit trails, responsible AI considerations, and bias risk in client contexts; formal framework experience is a bonus, applied judgment is not optional.

Employment

• Employment Type: Consultant
• Work Mode:
• Working Days: 5 days (Sat-Sun week off)
• Shift: Fixed Shift (9 hours) General Shift
• Shift Timings:
• Location: Bengaluru

Key Responsibilities

ML Foundation, Lifecycle & MLOps

• Own end-to-end ML pipeline design covering data preparation, feature stores, training, evaluation, deployment, and monitoring.
• Instrument and manage model drift data drift, concept drift, prediction drift with automated alerting and retraining workflows.
• Define and enforce ML engineering standards across the team: reproducibility, experiment tracking, code quality.

Agentic AI Engineering

• Architect and build multi-agent systems for complex, stateful, multi-step business workflows.
• Design agent communication patterns, tool registries, memory management, and failure recovery.
• Move beyond framework defaults understand what the orchestration layer is doing and when to build custom.
• Evaluate and select agentic frameworks based on problem requirements, not familiarity.

GenAI & Foundation Model Engineering

• Select, fine-tune, and serve foundation models appropriate to the problem; default to fit-for-purpose, not to the most capable or most expensive option.
• Manage prompt-to-model drift when a model version changes, understand what breaks and own the evaluation process.
• Design structured prompt engineering workflows with evaluation harnesses, not ad hoc iteration.

RAG & Knowledge Systems

• Design retrieval systems from requirements: choose between naive RAG, hybrid, Graph RAG, or knowledge graph-based approaches based on the problem, not the template.
• Select and configure vector databases appropriate to the retrieval use case; understand latency, scale, and cost trade-offs.
• Own retrieval quality end-to-end: relevance scoring, hallucination benchmarking, golden set evaluation.
• Progressively build competency and lead implementation in Graph RAG and structured knowledge retrieval.

LLM Evaluation & Quality

• Define and own the evaluation framework for all LLM and agentic AI solutions from unit evals through end-to-end system testing.
• Implement LLM Eval tooling (RAGAS, TruLens, DeepEval or equivalent) as a standard part of the delivery pipeline, not a post-hoc check.
• Establish golden datasets and ground truth benchmarks per domain; maintain and evolve them as models and prompts change.
• Report evaluation outcomes clearly to both technical teams and client stakeholders.

Technical Leadership & Delivery

• Set the engineering bar for the AI/GenAI team code reviews, architecture decisions, design standards.
• Work directly with pre-sales and solution design where technical depth is required in client engagements.
• Mentor mid-level engineers; translate ambiguous business requirements into clear engineering specifications.
• Represent the team credibly in technical client conversations across governance, architecture, and delivery.
• Graduates only (please ignore pursuing/drop out/12th or 10th pass).