Topic
All digests tagged AIE CODE summit
Suggested route
Move from earlier context through the available material to the newest signals. This route is a reading aid, not a difficulty ranking.
Arize's agent Alex evolved from a 50% hallucination prototype to a robust system. Key lessons include moving planning into custom harness code, treating large tool results as files, and using small markdown files for long-term memory.
Read article
How do you give AI agents real autonomy without catastrophic failure? John Feminella details five composable controls—from Agent Registry to Classifier use—to govern non-deterministic AI systems.
Read article
Runlayer secured 300+ agents by deploying a gateway to log every session and tool call. They found risks like prompt injection, data exfiltration, and cost blowups from model switches.
Read article
Optiver details how its on-prem AI platform handles the immediate, high-stakes P&L feedback loop of trading. Learn about the Inference Gateway, Meta Harness, and Agent Swarm architecture.
Read article
Building human-like voice agents requires focusing on the entire conversational flow, not just individual model components. Key strategies include measuring time to first audio, using async tool calls, and benchmarking the full stack with simulated callers.
Read article
The AI Engineer role is shifting from writing features to building self-improving loops. This requires building always-on agents that consume telemetry, run in durable sandboxes, and are continuously tested using advanced evaluation systems.
Read article
The fix involves 'native filtering' over a clustered vector index, which combines pre- and post-filtering to achieve 120ms P90 searches across 400 million vectors.
Read article
To ensure AI agent reliability, the speaker details a five-step regression loop. This process uses production traces as the source of truth, employing LLMs as judges to quantify subjective metrics and integrating automated gates into CI/CD.
Read article
Can AI understand a distressed credit? Harneet Kaur demonstrates how building a structured, multi-layered reasoning process forces top LLMs (Astra 6, Opus 5.5) to converge on a single probabilistic view of a distressed company.
Read article
Contour Asset Management shares how they measure AI ROI by running live A/B tests on analyst workflows. They focus on 'cost per useful insight' and integrating disparate data sources like AlphaSense and Yipit into unified platforms.
Read article
Arca's AI-native approach uses long-horizon agents to run a financial services business. The system manages complex, multi-day workflows by integrating signal systems, persistent memory, and shared spaces.
Read article
Securing autonomous AI agents requires a layered defense-in-depth approach. Key controls include one identity per agent, default deny egress (including DNS), and deterministic brokers for all tool calls.
Read article
Joseph Kim details how ChatGPT for Financial Services addresses the 'capability overhang' by integrating proprietary data, firm templates, and interactive tools, setting the stage for proactive, persistent AI agents.
Read article
Wells Fargo built a proprietary Foundation Model for transactions, pre-training a transformer on billions of raw customer transactions. This single backbone model improves fraud detection and streamlines multiple downstream financial tasks.
Read article
Most agent evaluations are easy to misread. To test AI agents like a scientist, you must design realistic tasks, run repeated tests, and focus on full traces and observability, because production is the final truth.
Read article
AI models are not the bottleneck for developer throughput. Analysis of 50,000 engineers shows that individual skill, team environment, and company culture are far more critical drivers than the latest AI model.
Read article
AI failures require a holistic, risk-based evaluation framework. Assess systems across Technical Strength, Deployment Controls, and Business Context using a 'build-break-fix' loop to ensure safety.
Read article
When public benchmarks are unavailable, the solution is to reverse the evaluation process: start with the known answer and generate the input. This shifts AI agent testing from slow expert review to automated, scalable code checks.
Read article
GEICO tackled the 'outer loop' bottleneck in SDLC by building an AI-native war room. This system uses agent fleets and structured rules (FleetKit) to accelerate major code modernization missions and automate decision-making.
Read article
A faster model only helps if the finished job gets faster. We test M3.1 Flash by measuring the total cost (time, compute, and manual fixes) of a complex research report, proving that speed must be balanced with accuracy and reliability.
Read article