Ravi Koka is teaching a machine how Wall Street thinks.
His North Side company, StockSnips, is home to an artificial intelligence-driven investing system that reads the financial news the way investors do, and then uses that information to pick stocks.
The technology operates on the idea that stock prices don’t move only when companies report quarterly earnings. They move every day in response to investor sentiment, which is guided by headlines, and the way companies are being discussed.
StockSnips is built to capture and interpret those positive and negative signals.
“That’s the big thing,” said Mr. Koka, CEO. “To prove that sentiment, or news-based sentiment, can actually impact the stock price.”
Mr. Koka and co-founder Viv Pennanti, created an algorithm that scans close to 50,000 news articles a day from about 25 news sources such as CNBC, Reuters and Motley Fool.
Read More
The system analyzes how individual companies are being talked about and derives a sentiment score based on whether the coverage is positive or negative. It then narrows the universe of about 3,000 stocks down to 350 names of high quality companies and overlays sentiment to determine the top 30 that are likely to outperform. Those stocks are held in a portfolio for a month before the model refreshes and picks another slate.
Investors have been putting real money behind the idea.
Roughly $30 million is now being traded using the StockSnips algorithm. That includes about $18 million in assets in the firm’s exchange-traded fund (ETF), which launched in April 2024 under the ticker symbol NEWZ and trades on the Nasdaq Stock Exchange, along with another $10 million that financial advisers trade in client accounts.
The market reacts
StockSnips was founded in 2016, and since its earliest days, the results have been encouraging.
The model had only ever been tested in a bull market since 2016, but at the end of 2021, the software determined on its own that the time had come to switch sectors. The models shifted out of tech and growth into stocks that were more in line with consumer staples, energy and some financial, Mr. Koka said.
The three-year annualized return for the S&P 500 at the end of December 2022 was 7.72% compared to 17.38% for StockSnips. In 2024, the algorithm delivered a total return of 15.16%.
But the market environment of 2025 has been a much tougher test.
It finished 2025 down 4.09% for the year, a stumble that Mr. Koka attributes to the unprecedented volatility of the current news cycle. Trade policy, tariffs and rapid-fire announcements from the Trump administration have whipsawed markets faster than headlines can keep up.
“Everything the Trump administration says and does, the stock market almost immediately reacts,” Mr. Koka said.
Some events, he added, simply defy modeling. Neither his system nor any other quantitative approach can reliably predict so-called “Black Swan” events, which are by definition, shocks that arrive without warning.
“Nobody exactly predicted what would happen on what they called Liberation Day,” he said. “The pace of news was not able to keep up. We are dependent on the news, and the news was not capturing this up-and-down roller coaster.”
Working out the AI kinks
For all the buzz surrounding artificial intelligence, adoption among financial advisers is still cautious and selective.
StockSnips, for example, currently has 15 financial advisers who pay for a license to use its model in client trading — an encouraging base, but hardly a stampede.
Many advisers are intrigued by the possibilities of AI, but prefer a watch-wait-and-see approach. At least for now, the risk of handing decision-making to a still-evolving technology outweighs the potential upside.
Matt Yanni, founder of Yanni & Associates Investment Advisors in Pine, falls squarely in the wait and see camp.
“It is possible we may utilize AI software in the future,” Mr. Yanni said. “But I’d rather their kinks be worked out with other companies, not ours.”
Mr. Yanni said he is aware that some advisers are already using AI tools to help construct portfolios or analyze investments, though the extent of that use isn’t clear. At his own firm, AI hasn’t yet made it into either the investment process or compliance functions.
More fundamentally, Mr. Yanni questions the idea that algorithms — no matter how sophisticated — can replace the human element at the core of financial advice.
“Personally, I do not believe AI will be able to fully replace financial advisers,” he said.
“It is quite common that many of the complex questions that we face require a considerable amount of existing knowledge of our clients as well as dialogue with them.”
A background rooted in tech
For Mr. Koka, StockSnips is a second act.
Long before AI and algorithms became buzzwords, he founded SEEC Inc., a technology company he took private in 1998 and later teamed up with IBM.
At its peak, SEEC employed about 200 people in offices near Pittsburgh International Airport before Mr. Koka sold the company and its headquarters moved to New Jersey.
Today, StockSnips is lean by comparison.
The company has a staff of six, focused on programming, quantitative modeling and refining the algorithm that powers its investing strategy.
The company has raised $4.5 million from investors — enough to build the technology and get it into the market, but not nearly enough to compete head-on with the asset management giants.
“If we were an established brand, we would have several billion dollars by now,” Mr. Koka said. “And that’s actually the next step. We’re going to partner with some of the bigger names and established brands.
“What we realize is you may have the technology, but building the brand and distribution will take time,” he said. “You have to raise a lot of money to take on BlackRock, and Vanguard and State Street and all of the big names.”






Comments
Welcome to Post-Gazette subscriber-only commenting community. Please review our community guidelines.