CTO @ PubNub | $134M raised | Series E | +1 Billion devices | 68 Patent claims | 25 Years

Seattle
Joined August 2008
Java remains practical, while Kotlin offers a cleaner JVM option. #java #springboot #kotlin
62
AI makes code feel disposable, but craftsmanship still matters. #coding #ai #craftsmanship
55
Run local RL simulations and see what the robot can learn. #reinforcementlearning #robotics #simulation
81
Stephen Blum retweeted
A #meatproxy copies and forwards AI-generated text, code, or other output without reading, understanding, or validating it. A human-shaped clipboard. If you're cool with that life, we've got some...tips... for you. Or you can follow along while CTO @stephenlb dives into how to reduce your AI spend each month while still churning out quality code. Read it here: bit.ly/4BooKI4
1
2
124
Sonnet rebuilt a Mac calculator app for just 36 cents. #sonnet #coding #mac
77
Zero percent testing does not eliminate real world prompt injection risks. #ai #security #promptinjection
44
Launch early, learn from users, and improve as you go. #gamedev #startups #iteration
1
4
121
AI makes software easier to build, increasing demand for engineers. #ai #software #engineering
1
3
125
Tool calling turns an LLM chatbot into a basic AI agent. #ai #agents #llm
1
1
3
109
Stephen Blum retweeted
Agents can build a whole app's UX now. What if stakeholders could review all the UX changes at once? To go from a V1 to a great app, you need to iterate until it's amazing. Use Decode, work together, and get there the fastest.
12
8
6
40
3,512
Stephen Blum retweeted
Take your LangChain agent public or share it with just those people who need it most with Blocks.ai. No replatforming, no messy middleware, no extra cost to you. How Blocks and #LangChain work together: bit.ly/4hNWzdY
1
4
4
85
Great collaboration matters more than certifications in our hiring process. #hiring #teamwork #communication
1
3
107
Handle collisions to prevent keys from overwriting existing data. #hashing #datastructures #coding
1
88
Learn C to understand computers closer to the machine. #programming #c #systems
1
1
199
Stephen Blum retweeted
Building #benchmarking and validation tools requires a clear vision for what you want to measure. Product Leader Cres Hay built a blocks.ai agent validation tool with this idea in mind. She defined the input and output criteria before creating anything. Then she built her agent and set it loose to measure agent effectiveness on Blocks. Learn how she did it, get her advice to new builders, and unpack product management in the #agentic age of #AI with her in our latest Blocks Builders series. bit.ly/4yoDggs
3
4
47
Choose Node.js for web backends, Python for broader opportunities. #nodejs #python #backend
1
2
94
AI uses less electricity to write code than a human at a desk: Human: 3 workdays (24 hrs) @ 135W = 3.24 kWh AI Pod: 10 mins @ 5,000W = 0.92 kWh Even drawing 40× the peak power, the AI finishes so fast it uses 3.5× less total energy, including time it takes to correct AI mistakes
1
3
92
Open source rights depend on the license, especially copyleft terms. #opensource #licensing #gpl
1
2
62
REA (reverse engineer anything) decompiled and rebuilt the mac notes app and calculator app for $0.63 in 19m10s
1
3
183
Scroll with two fingers on a flat mouse. #mouse #scrolling #tech
1
1
3
107