AI/ML Engineer

Job Title: AI/ML Engineer

Location: Bengaluru, KA (Hybrid - 3 Days/Week onsite)

Job Type: Fulltime

Role Definition

• Looking a genuine AI data scientist who can understand and use models based on context — not someone narrowly limited to a single tool or technique.
• Python expertise is non-negotiable — the candidate must know Python in and out; no learning on the job.
• Candidate should have breadth across the AI ecosystem: classical ML (e.g., XGBoost, SVM — data cleaning, feature engineering, training, cross-validation), NLP techniques (embeddings, text classification), and effective LLM usage (prompt engineering, training/fine-tuning small models/SLMs on corpus data).
• MCP was clarified as just one part of the broader context — does not want candidates restricted to it; they must be adaptable to pivot to NLP, model retraining, etc., without fumbling on fundamentals.

2. Production-Grade Mindset (Key Theme)

• Solutions are all production-grade and customer-facing — he stressed the significant gap between POC and production (scale and complexity are far higher).
• Candidates must come in understanding they are building customer-facing, production-grade solutions.

4. Ideal Background / Domain Fit

• Strong preference for candidates from product-based companies, ideally from the services org of a product company (e.g., own background at Philips Research servicing devices).
• Should understand concepts like log files and customer cases — directly relevant to the DEA/NSP customer-services domain.
• Must have worked with unstructured data (text, voice) rather than purely structured/numeric data (e.g., banking/financial candidates would struggle with the domain context and data type).

5. Top 5 Must-Have Skills

• Hands-on Python development (expert level; not team management)
• Agentic AI (confirmed as more important than GenAI)
• GenAI
• Understanding of the whole AI ecosystem / models
• Product-based service org understanding (how a service org works)