Notes from the
edge of the model.
Field notes on what actually breaks in production — agents, retrieval, evaluation, MLOps, and the career decisions nobody writes down. Longer arguments become papers; these are the rest.
353 posts
Digital twins — AI-powered virtual replicas of physical systems — are a $110B market by 2028.
What digital twins do: simulate a factory before building it. Test autonomous vehicle scenarios without real cars. Model a city's traffic patterns to optimize signal timing. Predict…
By Pranay Mahendrakar Read postIndia's AI market will hit $17B by 2027. Here's exactly where the jobs are and what they pay.
Top AI employers in India (2026): Big Tech India labs: Google, Microsoft, Amazon, Meta — hiring ML engineers at $50K-$150K equivalent. Cutting-edge research AND product work. Indian AI…
By Pranay Mahendrakar Read postAI Safety went from a niche academic concern to one of the hottest job markets in tech.
What happened: governments started regulating AI. Companies realized unsafe AI is a liability. Users demanded trustworthy systems. And several high-profile AI failures made the business…
By Pranay Mahendrakar Read postThe horizontal AI play (build a general platform for everyone) is incredibly competitive.
Vertical AI winners in 2026: Legal: Harvey, CaseText, EvenUp — AI that genuinely understands legal reasoning. Raising hundreds of millions. Healthcare: Viz.ai, PathAI, Tempus — AI for…
By Pranay Mahendrakar Read postVoice-first AI interfaces are about to change how millions of non-tech people interact with…
Think about who currently CAN'T use digital tools effectively: elderly people who struggle with touchscreens, illiterate populations, people with disabilities, and anyone who finds…
By Pranay Mahendrakar Read postData is the new oil? No. Data is the new CURRENCY.
The data supply chain for AI: data collection, data cleaning and annotation, data marketplace operations, synthetic data generation, data quality assurance, data privacy engineering…
By Pranay Mahendrakar Read postAI wearables are the next massive hardware + AI market.
Meta Ray-Ban smart glasses. Humane AI Pin (lessons learned). Apple Vision Pro's AI features. Rewind Pendant. Rabbit R1. The hardware is varied, but the trend is clear: AI is moving from…
By Pranay Mahendrakar Read postCybersecurity + AI is producing the highest average salaries in all of tech.
Average compensation for AI Security Engineers in 2026: US: $190K-$340K UK: £120K-£220K India: 35-80 LPA at top firms Remote: $150K-$280K Why so high? Two reasons: the threat landscape…
By Pranay Mahendrakar Read postClimate tech powered by AI is attracting more venture capital than any other vertical.
AI + Climate roles: Carbon Accounting ML Engineer (measuring and predicting emissions using satellite data and ML). Energy Grid Optimization Engineer (balancing renewable energy supply…
By Pranay Mahendrakar Read postPrompt Engineering is dying. Prompt Architecture is being born. And the salary difference is 3x.
A prompt engineer writes individual prompts. A prompt architect designs systems that generate, test, optimize, and manage prompts at scale. What a prompt architect builds: dynamic prompt…
By Pranay Mahendrakar Read postConstruction is a $13 trillion global industry and one of the LEAST digitized.
AI applications in construction 2026: drone-based site inspection with CV (detecting defects, measuring progress). Predictive scheduling that forecasts delays before they happen. Safety…
By Pranay Mahendrakar Read postBy 2028, I believe everyone will have a personal AI assistant that knows their preferences…
Not Siri. Not Alexa. Something fundamentally more capable — an AI that truly knows you. The technology is almost ready. Long-term memory systems. Personalization through fine-tuning on…
By Pranay Mahendrakar Read post