Multimodal AI: Five Use Cases That Actually Work (and Three That Do Not)
GPT-4V is out and everyone is building vision features. After testing it across real workflows, here is what ships well and what falls apart.
Fintech
Definition
Fintech coverage in this archive spans 7 posts from Feb 2017 to Dec 2023 and links technical decisions to margin, distribution, and execution durability. The strongest adjacent threads are ai, gdpr, and privacy. Recurring title motifs include ai, gdpr, fintech, and multimodal.
Key claims
- The posts consistently push for explicit unit economics and practical tradeoffs over narrative hype.
- Early posts lean on gdpr and engineering, while newer posts lean on ai and gdpr as constraints shifted.
- This topic repeatedly intersects with ai, gdpr, and privacy, so design choices here rarely stand alone.
Practical checklist
- Tie roadmap bets to measurable outcomes: cost, throughput, risk reduction, or revenue impact.
- Start with the newest post to calibrate current constraints, then backtrack to older entries for first principles.
- When boundary questions appear, cross-read ai and gdpr before committing implementation details.
Failure modes
- Treating technical strategy as branding instead of an operating constraint.
- Running broad experiments without clear stop conditions or budget discipline.
- Applying guidance from 2017 to 2023 without revisiting assumptions as context changed.
Suggested reading path
- Start here (current state): Multimodal AI: Five Use Cases That Actually Work (and Three That Do Not)
- Then read (operating middle): GDPR Week One: What Actually Happened
- Finish with (foundational context): GDPR Is an Engineering Problem, Not a Legal One
Related posts
- Multimodal AI: Five Use Cases That Actually Work (and Three That Do Not)
- AI Technical Debt Is Eating Your Codebase (You Just Cannot See It Yet)
- What I Learned Building AI Features Into a Fintech Product
- GDPR Week One: What Actually Happened
- GDPR for Engineers: What We Actually Built at a Fintech Startup
- Event Sourcing in Practice: What I Got Right and Wrong
- GDPR Is an Engineering Problem, Not a Legal One
References
AI Technical Debt Is Eating Your Codebase (You Just Cannot See It Yet)
AI features create a new species of technical debt that hides in prompts, data pipelines, and model versions. By the time you notice it, the cleanup bill is brutal.
What I Learned Building AI Features Into a Fintech Product
Building AI features at a fintech infrastructure company taught me that the hard part isn't the model. It's defining quality, handling failures gracefully, and resisting the urge to ship a demo as a product.
GDPR Week One: What Actually Happened
GDPR went live on May 25th. Here's what the first week looked like from inside a fintech company -- the scrambles, the surprises, and the things we got right.
GDPR for Engineers: What We Actually Built at a Fintech Startup
Eleven days before the GDPR deadline, here's the technical implementation work we did at the fintech startup — data mapping, consent storage, erasure pipelines, and the backup problem nobody warns you about.
Event Sourcing in Practice: What I Got Right and Wrong
Lessons from building event-sourced systems at the fintech startup -- the patterns that held up, the modeling mistakes that bit us, and the operational realities nobody warns you about.
GDPR Is an Engineering Problem, Not a Legal One
We're 15 months from GDPR enforcement. Here's the technical checklist I'm working through at the fintech startup — data inventory, consent, deletion, and everything else engineering actually has to build.