Hi, my name is
Tanish Rana.
I build systems to solve problems that matter.
I'm drawn to technical challenges where the solution isn't obvious and the stakes are high. Whether it's ML systems that need to be right, infrastructure that can't go down, or products that have to scale, I want to work on problems where engineering matters.
Get In TouchAbout Me
Hey! I'm Tanish, a Computer Science graduate student at NYU with a passion for building intelligent systems that solve real-world problems. My journey in tech has taken me from developing analytics platforms for pharmaceutical companies to developing ML and software systems for naval operations.
I specialize in full-stack development and machine learning, with hands-on experience building cloud-native applications, NLP-based automation tools, and scalable data analytics platforms. Whether it's architecting a collaborative workspace at scale or automating DevOps workflows with natural language processing, I'm driven by creating solutions that make a measurable impact.
I'm particularly interested in systems that sit at the intersection of machine learning and production engineering: where models need to be both intelligent and reliable, where infrastructure needs to handle real scale, and where the technical decisions directly determine whether something works or breaks. I'm looking for teams solving hard problems that matter.
Here are some technologies I work with:
- Python
- C++
- TypeScript
- SQL
- Rust
- TensorFlow
- PyTorch
- Django
- React
- Node.js
- AWS
- Docker
- Kubernetes
- R
- Supabase
Education
Where I've Worked
Business Analyst (Sales Operations) @ ProcDNA Analytics
- Owned and delivered 100+ sales operations dashboards and ad-hoc analysis deliverables for a S&P 500 pharmaceutical client, building SQL pipelines to validate data and eliminate recurring integrity issues
- Automated Attainment, Contest, and Incentive Compensation reporting workflows using Excel and VBA, reducing manual effort by 75% and improving operational efficiency
- Conceptualized and developed a secure, modular ICD Code Governance Platform (Python, SQL, VBA) with audit trails, cutting approval cycles by 50%; commercialized and sold for $20K to client
- Partnered directly with client stakeholders to define requirements, design scalable analytics tools, and translate data into actionable insights
Research Intern @ Indian Navy - WESEE
- Developed TensorFlow-based anomaly detection models utilizing unsupervised learning algorithms, achieving 92% accuracy in identifying irregular AIS ship transmissions
- Streamlined confidential real-time datasets through advanced data cleaning, improving reliability and accuracy of deployed anomaly detection models
- Contributed to combat system software in C++ with Kafka middleware for real-time sensor data streaming, and developed PyQt dashboards for mission-critical operational visualization
- Developed unsupervised ML models for HR attrition prediction, achieving 97% accuracy
Personal Projects
- A full-stack collaborative workspace platform with React frontend and Django REST backend supporting 200+ concurrent users. Features secure Auth0 authentication, AWS services integration, and automated CI/CD deployment with 99% uptime.
- Cloud-native NLP-based DevOps control plane using AWS Lambda and Amazon Lex to translate natural language commands into automated CI/CD workflows. Reduced deployment time by 60% with event-driven orchestration and achieved 99.9% availability.
- Sentiment-driven trading models using NLP on 82K+ financial news headlines. Built ML classifiers and LSTM networks achieving 93% test accuracy and grew a simulated $100K portfolio to $150K (+49%) through backtested strategies.
Recent Posts
A deep dive into parameter-efficient fine-tuning methods, comparing LoRA and QLoRA against full fine-tuning with real cost analysis from production deployments.
LLMsMachine LearningMLOpsCost AnalysisExploring the evolution of transformer context windows, from the original 512 tokens to million-token models, with practical insights on cost-performance trade-offs.
LLMsMachine LearningArchitectureCost AnalysisReal-world security incidents from deploying LLM-powered systems, including a GitHub Copilot vulnerability that reached production and practical defense strategies.
AI SafetySecurityLLMsDevOps
Memorable Reads
The Design of Everyday Things
by Don NormanA foundational text on user-centered design. Norman's principles of affordances and feedback fundamentally changed how I think about building intuitive systems.
Refactoring UI
by Adam Wathan & Steve SchogerPractical design tactics for developers. This book taught me that good design isn't about talent—it's about following systematic principles that work.
Thinking, Fast and Slow
by Daniel KahnemanNobel Prize-winning insights into cognitive biases and decision-making. Essential for understanding how humans process information and make choices under uncertainty.
Influence: The Psychology of Persuasion
by Robert CialdiniThe science behind why people say yes. Understanding these principles has improved everything from product design to stakeholder communication.
Atomic Habits
by James ClearCompound effects of 1% improvements. This systematic approach to habit formation has been invaluable for maintaining consistency in learning and execution.
The Pragmatic Programmer
by Andrew Hunt & David ThomasTimeless principles that transcend languages and frameworks. The emphasis on craftsmanship and continuous improvement has shaped my engineering philosophy.
Designing Data-Intensive Applications
by Martin KleppmannThe definitive guide to distributed systems. This book explains the tradeoffs in database design, streaming, and distributed computing with exceptional clarity.
Clean Architecture
by Robert C. MartinPrinciples for building maintainable systems. Martin's SOLID principles and dependency rules have guided my approach to structuring scalable applications.
Deep Learning
by Ian Goodfellow et al.The comprehensive ML bible. Dense but thorough coverage of neural networks, optimization, and generative models. Essential for serious ML practitioners.
Hands-On Machine Learning
by Aurélien GéronPractical ML with Scikit-Learn and TensorFlow. Bridges theory and implementation beautifully, with clear examples and production-ready code patterns.
A Random Walk Down Wall Street
by Burton MalkielThe efficient market hypothesis explained. A reality check on market prediction and a compelling case for evidence-based investing strategies.
The Intelligent Investor
by Benjamin GrahamWarren Buffett's recommended investing guide. The concept of "Mr. Market" and margin of safety fundamentally shaped my approach to valuation and risk.
Flash Boys
by Michael LewisExposes high-frequency trading dynamics. Eye-opening look at market microstructure and how technology advantages translate to financial returns.
Sapiens
by Yuval Noah HarariA sweeping history of humankind. Harari's analysis of cognitive, agricultural, and scientific revolutions provides context for understanding technological change.
The Black Swan
by Nassim Nicholas TalebImpact of rare, unpredictable events. Taleb's critique of Gaussian models and emphasis on antifragility changed how I think about risk and system design.
Superintelligence
by Nick BostromPhilosophical examination of AGI risks. A rigorous analysis of control problems and existential risks that every AI practitioner should consider.
Zero to One
by Peter ThielCreating value through innovation, not competition. Thiel's contrarian thinking and monopoly thesis offer a refreshing perspective on building technology companies.
What's Next?
Get In Touch
I'm currently looking for new opportunities and interesting conversations. Whether you have a question, want to collaborate, or just want to connect, I'd love to hear from you!