Predikly hiring for Gen AI & ML Architect.

Job Location : Shivajinagar, Pune

Mode : Hybrid

ROLE PURPOSE

As Gen AI & ML Architect, you will own the end-to-end design, development, and delivery of production-grade Gen AI and Machine Learning solutions for Predikly's clients. You will lead a team of 5-10 AI/ML engineers, drive technical architecture decisions, and serve as a hands-on contributor writing code, reviewing models, and solving complex problems alongside your team. This is a player-coach role demanding both individual technical depth and team leadership maturity.

KEY RESPONSIBILITIES

1. Team Leadership & Mentoring

- Lead, manage, and mentor a team of 5-10 Gen AI & ML Developers, driving accountability, growth, and technical excellence.
- Conduct code reviews, design reviews, and sprint ceremonies; set engineering standards and best practices.
- Collaborate with product managers, delivery leads, and client stakeholders to define roadmaps and prioritise backlogs.
- Foster a culture of experimentation, continuous learning, and engineering rigour within the AI/ML team.

2. Agentic AI Development

- Design and build production-grade agentic AI systems for real customer use cases — including multi-agent orchestration, autonomous reasoning pipelines, and tool-use patterns. Architect and implement solutions using frameworks such as LangChain, LangGraph, AutoGen, CrewAI, and custom agent orchestration layers.
- Define agent memory, planning, reflection, and tool-call patterns aligned with real-world reliability requirements.
- Ensure agents meet production-readiness criteria: latency SLAs, cost budgets, fallback handling, and observability.

3. Gen AI & LLM Engineering

- Work hands-on with leading LLM providers: OpenAI (GPT-4o, o1), Anthropic Claude, Google Gemini, and open-source models (LLaMA, Mistral).
- Design and implement Retrieval-Augmented Generation (RAG) pipelines with hybrid search, re-ranking,



and context compression.
- Engineer advanced prompting strategies: chain-of-thought, structured output, few-shot, system persona, and dynamic routing.
- Build vector search & embedding pipelines using tools such as Pinecone, Weaviate, Qdrant, pgvector, and FAISS.
- Evaluate, fine-tune, and benchmark models; manage prompt versioning and model drift monitoring.

4. Machine Learning Engineering

- Design and deliver end-to-end ML pipelines: data ingestion, feature engineering, model training, evaluation, and deployment.
- Apply ML techniques including classification, regression, clustering, NLP, time-series forecasting, and transformer-based architectures.
- Leverage MLflow, Weights & Biases, or Kubeflow for experiment tracking, model registry, and pipeline orchestration.
- Implement monitoring and feedback loops for model drift, data drift, and retraining triggers in production.

5. Cloud Architecture & Infrastructure

- Architect and deploy AI/ML workloads on AWS, Azure, and GCP using managed AI services and custom containerised deployments.
- Use AWS SageMaker, Azure ML, Google Vertex AI, Lambda, ECS, and serverless patterns to build scalable AI services.
- Design secure, cost-optimised cloud architectures; apply IAM, VPC, secrets management, and cost tagging best practices.
- Set up CI/CD pipelines for ML workflows using GitHub Actions, Azure DevOps, or AWS CodePipeline.

6. Full-Stack AI Application Development

- Contribute to and guide development of AI-powered web applications using React, Angular,



Next.js on the frontend and Node.js on the backend.
- Design and manage databases including PostgreSQL, MySQL (relational) and MongoDB, DynamoDB, Redis, Elasticsearch (NoSQL).
- Build REST and GraphQL APIs that surface AI capabilities to end-user products; ensure API security, versioning, and documentation.
- Integrate AI/ML model outputs into product UIs with appropriate UX patterns for streaming, latency masking, and error handling.

7. Client Orientation & Stakeholder Engagement

- Serve as the primary technical interface for client engagements — leading discovery workshops, solution walkthroughs, and sprint demos to build confidence and trust with business and technical stakeholders.
- Translate complex AI/ML concepts, architectural trade-offs, and model limitations into explicit, business-relevant language that non-technical clients and product owners can act on. • Collaborate with client teams to identify high-impact AI use cases, define success metrics, and ensure delivered solutions align with stated business outcomes — not just technical specifications.

REQUIRED QUALIFICATIONS & EXPERIENCE

Experience

- 8–10 years of total software/AI engineering experience.
- Minimum 3 years in hands-on Generative AI & ML project delivery in production environments.
- Proven experience leading a team of 5-10 AI/ML developers with direct responsibility for delivery, quality, and team development.
- Demonstrated track record of shipping agentic AI systems to real customers — not just PoCs or prototypes.
- Experience working in an Agile/Scrum delivery model, including sprint planning, stand-ups, retrospectives, and stakeholder demos.

Education

- B.E. / B.Tech / M.Tech in Computer Science, AI/ML, Data Science, or a related technical discipline.

Interested please share resume on [email protected]

📌 AI/ML Solution Architect (Pune)
🏢 Predikly Technologies
📍 Pune