- Developed quantitative models in Python and SQL to analyze capital flows, market positioning, and sector performance — supporting investment research and strategic decision-making at an institutional level.
- Designed financial dashboards and automated reporting systems to monitor asset performance, risk metrics, and exposure across multiple market segments in real time.
- Integrated live market data APIs into analytical workflows, improving data accuracy and enabling significantly faster research turnaround for senior stakeholders.
- Collaborated with cross-functional teams to translate complex investment questions into structured analytical frameworks and clear, actionable insights.
- Conducted rigorous scenario analysis and stress testing simulations to quantify downside risk under a range of adverse market conditions.
- Prepared polished research summaries, analytical presentations, and executive briefing notes for senior team members and portfolio decision-makers.
- Led ad-hoc research initiatives on sector trends, market dislocations, and structural macro shifts impacting capital allocation strategy.
- Applied advanced NLP techniques (spaCy, TextBlob, transformers) to extract and quantify sentiment from financial filings and market news, boosting trend analysis accuracy by 20%.
- Automated data aggregation and preprocessing pipelines using Python and SQL, cutting data preparation time by 60% and dramatically improving dataset integrity.
- Utilized relational databases to manage and query large financial datasets, enhancing retrieval efficiency across research workflows.
- Presented data-driven research findings using Matplotlib and Seaborn visualizations at internal seminars, earning recognition from faculty supervisors.
- Built an end-to-end ML pipeline forecasting next-day stock price direction by combining real-time news sentiment with historical OHLCV data
- Extracted polarity scores using TextBlob and engineered sentiment-derived features that consistently outperformed price-only baseline models
- Trained XGBoost with time-series cross-validation, achieving 67.52% test accuracy under simulated live market conditions
- Designed the pipeline to be modular and scalable — clean separation between data ingestion, feature engineering, model training, and evaluation
- Deployed to GitHub with SSH authentication; documented for reproducibility and future extension to additional tickers
- Architected a full-stack gym management platform using React, Node.js, and MySQL — handling class scheduling, membership tracking, and member engagement
- Leveraged AWS RDS for secure, scalable cloud storage with automated backups and high-availability configuration
- Built and maintained CI/CD pipelines via Jenkins enabling zero-downtime deployments across development and production environments
- Integrated AWS SNS for real-time push notifications — improving member communication and reducing no-show rates
- Designed a responsive UI with role-based access control for admins, trainers, and members with distinct permission levels
- Reduced manual admin workload by automating scheduling conflicts, payment reminders, and attendance reporting
- Built a systematic stock screening engine combining financial health metrics, valuation ratios, and market momentum using Pandas and Scikit-learn
- Leveraged NumPy to benchmark individual stocks against sector peers — improving analytical decision accuracy by 20% over manual screening methods
- Incorporated fundamental indicators including P/E, debt-to-equity, revenue growth, and free cash flow into a composite scoring model
- Applied clustering algorithms to group stocks by risk/return profile, enabling better portfolio diversification decisions
- Provided a structured, data-driven framework for value investing grounded in quantitative fundamentals and competitive positioning
- Developed an 85%-accurate sentiment classification model using Python and NLTK to analyze large volumes of social media text at scale
- Built a full NLP pipeline covering text preprocessing, tokenization, stop-word removal, stemming, TF-IDF feature engineering, and multi-classifier evaluation
- Designed interactive Tableau dashboards visualizing sentiment trends over time — enabling non-technical stakeholders to extract actionable insights without SQL
- Tested multiple classifiers including Naive Bayes, Logistic Regression, and SVM — selected the best performer based on F1 score across diverse social datasets
- Automated the data ingestion and processing pipeline to handle new batches of social data with minimal manual intervention
- Built a real-time player detection and tracking system using Python and OpenCV, enhancing detection accuracy by 35% through custom fine-tuning and augmentation
- Applied transfer learning on a pre-trained object detection model, adapting it to sports-specific scenarios with a custom labeled dataset
- Generated spatial heatmaps of player movement across the field — providing coaches with tactical positioning insights previously only available via manual review
- Optimized inference latency for near-real-time performance on standard hardware without requiring GPU clusters
- Designed the output pipeline to export frame-by-frame tracking data to CSV for downstream statistical analysis
- Built an automated financial planning model to track expenses, optimize savings rates, and forecast investment growth across multiple market scenarios
- Implemented a capital allocation framework illustrating the compounding effect of incremental savings on a 10-year portfolio horizon
- Executed risk-adjusted investment analysis covering wealth preservation strategies — presenting CAGR, ROI, and inflation-adjusted returns via custom Excel and Python dashboards
- Modeled three distinct scenarios (conservative, moderate, aggressive) to give users a clear picture of how savings behavior impacts long-term outcomes
- Incorporated Monte Carlo simulation to stress-test projections against market volatility and sequence-of-returns risk
- Organized monthly poker tournaments merging competitive gameplay with structured lessons in probability, expected value, and financial decision-making, bridging the felt and the markets.
- Taught members strategic gameplay, bankroll management, hand range analysis, and behavioral reading, skills that directly mirror position sizing, fundamental analysis, and investor psychology.
- Built a campus community at the crossroads of game theory and financial markets, fostering analytical sharpness and interpersonal edge among students pursuing finance careers.
- Recruited and mentored new members, growing club membership and establishing a culture of intellectual curiosity, disciplined thinking, and calculated risk-taking.
Poker — I've been playing seriously for years. It's not a hobby to me, it's a thinking system. Pot odds, range construction, tilt control. The same mental muscles you need at a trading desk or in a high-stakes client meeting.
Finance & Markets — I follow macro developments, equity research, and market structure closely. I read earnings calls, track Fed commentary, and spend time understanding how capital actually moves. Not because I have to. Because I genuinely find it interesting.
Building things — Whether it's a side project, a financial model, or a dataset I found interesting, I like the process of turning a rough idea into something that actually works. Most of my best learning has come from breaking things and figuring out why.
New York — Being in this city matters to me. The access to ideas, industries, and people who are operating at a high level every day is something I don't take for granted.
I'm Guru Akhil Tavva.
I'm someone who's constantly building. Not loudly. Not to prove a point. Just because that's who I am.
I like growth. I like putting myself in rooms that challenge me and make me better. I learn fast because I'm genuinely curious. I adapt quickly because I don't get attached to comfort. When I see an opportunity, I don't overthink it. I prepare, I show up, and I give it everything.
I've always believed that consistency and effort compound over time. You don't need noise. You need discipline. You need patience. You need the courage to keep going when nothing is guaranteed.
Poker shaped a big part of how I think. It taught me how to stay calm when the pressure rises. How to make decisions without having all the information. How to separate emotion from logic. How to accept losses without losing confidence. And how to recognize the right moment to take a real shot.
That mindset carries into everything I do.