Hila
Vianai Systems · Product Design
Conversational AI for enterprise financial analysis.
The Challenge
“Hallucinations Cost Millions”
In finance, a wrong number isn’t just a bug; it’s a liability. The fundamental roadblock was that standard LLMs (like ChatGPT) are prone to “hallucinations.” Our target users—financial analysts—needed absolute certainty, not just conversational fluency. We had to design an interface that prioritized verification over magic.
The Solution
Trust through transparency. We moved away from the “black box” chat interface to a system that explicitly “shows its work.”
1. Calculation Transparency (Visualizing the Logic)
- Concept: Analysts don’t trust a summary without seeing the math.
- Execution: We designed interactive data tables that bridge the gap between natural language summaries and raw enterprise data. Users can toggle between narrative, data grids, and visual charts to cross-reference every claim.
2. Direct Source Attribution
- Concept: “Calculated Trust.” Different users need different levels of proof.
- Execution: We implemented a multi-layered verification system.
- The Reasoning Tab: Reveals the specific arithmetic (e.g., CAGR formulas).
- Side-by-Side View: Clicking a data point opens the original source PDF with the exact relevant section highlighted.
3. The “Streaming Thought” UI
- Concept: Complex financial analysis takes time (5-10s), causing user anxiety (“Did it crash?”).
- Execution: Instead of a static spinner, we exposed the AI’s “thinking process” in real-time steps (e.g., “Searching 10-K…” → “Extracting Revenue…”). This managed wait time and built a mental model of competence.
Beyond Financials: The Probabilistic Shift
My work on hila (and previously on Kinect) has centered on the same fundamental shift in computing: The move from deterministic to probabilistic interfaces.
- Deterministic (Traditional): You click a button; a specific action happens.
- Probabilistic (AI/Gesture): The system “guesses” your intent based on noisy input or ambiguous prompts.
My Strategy for the AI Era:
- Smooth the Jitter: Don’t show the user the raw machine uncertainty; provide a filtered, confident UI.
- Affordances Matter: Empty chat boxes cause “Blank Canvas Anxiety.” Always provide a “hand to hold” through suggestions and visible constraints.
- Reliability > Magic: A feature that is “cool” but fails 20% of the time is a liability. I design for the 99% use case.
The Results
The “accuracy over impressiveness” approach validated the product immediately, turning an experiment into a revenue driver.
- Growth: 0 to 4,000 users in 4 months.
- Revenue: Reached $1M ARR in 4 months.
- Scale: Design team grew from 2 to 4; engineering team scaled from 5 to 50.
User Testimonials
“Connect it to your data sources (no need to build a data lakehouse) and it’ll just start answering your questions. After a few thumbs up/down and real-time reinforcement learning it’ll get it right. Almost magic.”
— Boris Evelson, VP at Forrester Research
“hila vastly improves my research process. I can rapidly search 10-Ks and earnings calls to find if anything related to my theses are hidden inside… More importantly, I can do it quickly without wasting time skimming irrelevant topics or pinpointing key words.”
— Mike Ostroff, Investment Analyst at Maverick Capital
“Being able to monitor and improve LLM performance is critical to unlocking the true power of gen AI. Vianai’s hila Enterprise provides clients a platform to safely and reliably deploy any large language model (LLM), optimized and fine-tuned to speak to their systems of record.”
— Ravi Kumar S, CEO of Cognizant
“Vianai is helping customers innovate by bringing its hila agents to Google Cloud. Leveraging the power of Gemini models, these solutions allow businesses to easily deploy sophisticated analytics without technical expertise, unlocking value from their data faster and more effectively.”
— Kevin Ichhpurani, Corporate VP at Google Cloud
Reflection
hila taught me that the best AI products feel effortless not because the AI is sophisticated, but because the design earns user trust.
We succeeded by making the most interpretable AI on the market. Every design decision—from source highlights to honest uncertainty to streaming thought processes—served the goal of making a powerful, unpredictable technology feel reliable and controllable.