Yourstory iconYourstorySep 5, 2026 ~4 min source read

Puneet Chandok: Learning the new tech is a “one-on-one full contact body sport” — a practical guide for workers and enterprises

Microsoft India & South Asia president Puneet Chandok lays out how knowledge workers and companies should adapt: learn by doing with tools daily, preserve human judgement, rebuild workflows for continuous model execution, and deliver contextualised intelligence with strong data guardrails.

Share this story

Send the public story page.

Useful takeaways from this story.

Human skills — taste, judgment, curiosity, empathy — become the differentiators as automation handles reasoning and execution.

Enterprises must redesign legacy workflows for continuous token-based execution instead of layering models on broken systems.

# Quick summary

# What Chandok recommends for knowledge workers Start using the tools directly every day. Chandok warns against passive learning—watching theoretical videos or delegating discovery to teams. He wants senior and middle managers to interact with the tools themselves so they develop practical muscle memory for when systems make recommendations and automated agents take action.

He frames the worker role as a barbell: technical capability on one end and distinctively human strengths on the other. As the technology takes on reasoning and execution, human qualities such as taste, judgment, curiosity, and empathy increase in value. The takeaway for individuals: build hands-on tool fluency while cultivating higher-order judgement and sense-making.

# What Chandok recommends for enterprises Companies need a clear ROI path before broad adoption. Chandok urges firms to stop pasting models over broken operations. Instead, redesign workflows for continuous, 24/7 token-based execution so model outputs can be actioned reliably at scale.

Operational changes include mapping where automated agents will execute, ensuring data pipelines support always-on requests, and rethinking approval and oversight to match faster execution cycles. Leaders should identify frontier professionals who can combine agent automation with sound human judgement and place them where decisions matter.

# Implications for startups and vendors Enterprise-focused startups can't succeed by merely renting frontier models and repackaging them. Chandok says they must deliver contextualised intelligence that understands a customer's domain and enforce strict data security guardrails. That combination—domain awareness plus data control—creates defensible value for customers who need secure, relevant automation rather than generic model outputs.

# Short practical checklist

  • Leaders: spend time daily using the tools that will touch your teams. Teach by doing.
  • Managers: embed practical tool use into team routines and decision checkpoints.
  • Ops teams: audit legacy workflows and design for continuous token execution and reliable data flows.
  • Vendors/startups: productise contextualised intelligence and operationalise strong data security as part of the offering.

# Why this matters now Chandok points to India's fast rate of frontline adoption and a rising cohort of professionals who pair automation with human judgement. For countries and companies moving quickly, the risk is rushing deployment without operational redesign and human-centred practices. The guidance on hands-on learning, workflow redesign, and contextualised, secure product design offers a concise path for turning tool-driven capability into usable value.

# Bottom line

More context around this story.

Loading more related stories...

Keep reading in the app

Open the app view to save this story, compare related coverage, and continue from the same source.

Open in app